Negotiation support device, negotiation support method, and program

JP7936643B1Active Publication Date: 2026-10-16A TO J INC
View PDF 5 Cites 0 Cited by

Patent Information

Application Number
JP2026149968
Authority / Receiving Office
JP · JP
Patent Type
Patents
Current Assignee / Owner
Priority Date
2026-01-20
Filing Date
2026-07-07
Publication Date
2026-10-16
Estimated Expiration
2046-07-07

AI Technical Summary

Benefits of technology

【0080】 本発明による交渉支援装置によれば、適切なAIエージェントを構築し、申立人と被申立人との交渉を支援できる。

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 0007936643000001_ABST
    Figure 0007936643000001_ABST
Patent Text Reader

Abstract

This invention provides a negotiation support device, negotiation support method, and program to assist in negotiations between the petitioner and the respondent. [Solution] A negotiation support device 5 that constructs an appropriate AI agent to support negotiations comprises a claim acquisition unit 121 that acquires claim information having claim content information, a claim storage unit 131 that stores claim information, and a negotiation support unit 534 that supports negotiations between the claimant and the respondent. The negotiation support unit 534 comprises one or more role-acting AI agents that support negotiations on behalf of the claimant, respondent or a third party, or agent means 5341 that realize one or more support AI agents that support the claimant, respondent or a third party in negotiations.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to a negotiation support device and the like that support negotiations between a petitioner and a respondent. Here, the petitioner may include a representative of the petitioner or a supporter of the petitioner. Also, the representative of the petitioner and the supporter of the petitioner may be considered separately from the petitioner. The respondent may include a representative of the respondent or a supporter of the respondent. Also, the representative of the respondent and the supporter of the respondent may be considered separately from the respondent.

Background Art

[0002] Conventionally, there has been a system that allows card issuers or other electronic payment providers to obtain quick, desirable and efficient resolutions to disputes between customers and member merchants (see Patent Document 1).

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] However, conventional technologies have not been able to appropriately utilize AI agents to support negotiations between the petitioner and the respondent regarding a claim. Furthermore, conventional technologies have not made it easy to construct an appropriate AI agent in a negotiation support device that assists negotiations between the petitioner and the respondent. Note that an appropriate AI agent may be an AI agent instructed by the user. The user is a party involved in the negotiation. Parties involved are, for example, the petitioner or the respondent. A third party is someone other than the petitioner or the respondent who supports the negotiation. A third party is, for example, an agent, mediator, secretariat, or supporter. An agent is, for example, an agent for the petitioner or an agent for the respondent. A mediator is, for example, a judge or arbitrator. A secretariat is, for example, a person who handles consultations, procedural management, and other administrative tasks. A supporter is a person who supports negotiations or agreement formation between the parties. A third party is, for example, a lawyer.

[0005] Furthermore, conventional technologies failed to perform appropriate processing on the negotiation support device at the appropriate time. Also, conventional technologies failed to utilize the AI ​​agent at the appropriate time.

[0006] Furthermore, conventional technology was unable to propose a settlement plan for reaching a negotiated agreement. Also, conventional technology was unable to propose a settlement plan for reaching a negotiated agreement at an appropriate time.

[0007] Furthermore, with conventional technology, the applicant's AI agent could not assist in negotiations. Also, with conventional technology, the applicant's AI agent could not assist in negotiations at the appropriate time.

[0008] Furthermore, with conventional technology, the respondent AI agent could not assist in negotiations. Also, with conventional technology, it was not possible to use the respondent AI agent to assist in negotiations at the appropriate time.

[0009] Furthermore, conventional technologies were unable to provide negotiation support through third-party AI agents. Additionally, conventional technologies were unable to provide negotiation support through third-party AI agents at the appropriate time.

[0010] Furthermore, with conventional technology, it was not possible to support the petitioner using the petitioner support AI agent. Also, with conventional technology, it was not possible to support the petitioner using the petitioner support AI agent at the appropriate time.

[0011] Furthermore, with conventional technology, it was not possible to support the respondent using the respondent support AI agent. Also, with conventional technology, it was not possible to support the respondent using the respondent support AI agent at the appropriate time.

[0012] Furthermore, conventional technologies were unable to support third parties assisting in negotiations using third-party AI agents. Also, conventional technologies were unable to support third parties using third-party AI agents at the appropriate time.

[0013] Furthermore, conventional technology could not assess neutrality in negotiations. Also, conventional technology could not output the neutrality assessment results at an appropriate time.

[0014] Furthermore, with conventional technology, it was not possible to conduct negotiations with the respondent using an AI agent at the appropriate time.

[0015] Furthermore, with conventional technology, when assisting negotiations between the petitioner and the respondent regarding a claim, it was not possible for the parties (e.g., the petitioner or the respondent) and the AI ​​agent to negotiate by telephone.

[0016] Furthermore, with conventional technology, when assisting negotiations between the petitioner and the respondent, if the respondent did not answer the call from the AI ​​agent, the AI ​​agent was unable to call the respondent at the appropriate next time.

[0017] Furthermore, conventional technology was unable to adequately answer questions from the petitioner or respondent regarding negotiation support devices that assist in negotiations between the petitioner and the respondent.

[0018] It should be noted that the above-mentioned problems in the prior art are not necessarily limited, and it may be considered that, along with or in lieu of the above-mentioned problems, there are problems corresponding to any of the problems or effects described herein, or other problems or effects that can be understood from this specification. [Means for solving the problem]

[0019] The negotiation support device of the first invention is a negotiation support device that supports at least a part of the phases from the start of negotiations to the end of negotiations or the completion of the performance of the agreed-upon terms regarding a dispute between a petitioner and a respondent, and comprises a petition acquisition unit that acquires petition information having petition content information relating to the content of the petition, a petition storage unit that stores petition information, and a negotiation support unit that supports negotiations between the petitioner and the respondent, wherein the negotiation support unit comprises agent means that realize one or more role-acting AI agents which are AI agents that support negotiations on behalf of the petitioner or respondent or a third party supporting the negotiation, and which perform proxy processing which is processing for interacting with a person in another position on behalf of the petitioner or respondent or a third party, or one or more support AI agents which are AI agents that perform support processing to acquire information to be output to the petitioner or respondent or a third party in order to support the petitioner or respondent or a third party in the negotiation.

[0020] This configuration enables the creation of an appropriate AI agent in a negotiation support device that assists negotiations between the petitioner and the respondent, thereby supporting the negotiations.

[0021] Furthermore, the negotiation support device of the second present invention further comprises a prompt management unit storing a settlement proposal creation prompt that is a prompt including an instruction to create a settlement proposal for a dispute, in addition to the first invention, wherein the agent means provides, in the negotiation, one or more types of information selected from negotiation information input by a claimant, negotiation information input by a respondent, negotiation information input by a third party, or generated information that is information output by any of the AI agents, and the settlement proposal creation prompt to a generative AI, acquires a settlement proposal from the generative AI, and comprises a settlement proposal presenting means for outputting the settlement proposal to one or more involved parties among the claimant, the respondent, or the third party.

[0022] With this configuration, a settlement proposal for reaching an agreement in negotiation can be presented.

[0023] Furthermore, in the negotiation support device of the third present invention, in addition to the second invention, the settlement proposal presenting means outputs the settlement proposal when a settlement proposal creation condition is satisfied.

[0024] With this configuration, a settlement proposal for reaching an agreement in negotiation can be presented at an appropriate timing.

[0025] Furthermore, the negotiation support device according to the fourth aspect of the present invention, in any one of the first to third aspects of the invention, further comprises a prompt management unit storing a claimant AI agent configuration prompt which is a system prompt for configuring the claimant AI agent that is an AI agent supporting negotiation on behalf of the claimant, and the agent means provides the claimant AI agent configuration prompt to generative AI, provides claim information to the claimant AI agent, and executes claimant proxy processing which is processing for interacting with the respondent, a third party, or another AI agent on behalf of the claimant. The claimant proxy processing includes proxy processing of receiving negotiation information from the respondent or the third party, or acquiring interaction information or generation information output from an AI agent other than the claimant AI agent, passing the negotiation information, the interaction information, or the generation information to the claimant AI agent, acquiring a response from the claimant AI agent, and outputting negotiation information based on the response to the respondent, the third party, or the AI agent other than the claimant AI agent.

[0026] With this configuration, negotiation can be supported by the claimant AI agent.

[0027] Furthermore, in the negotiation support device according to the fifth aspect of the present invention based on the fourth aspect, the agent means executes the claimant proxy processing when a claimant agent cooperation condition is satisfied.

[0028] With this configuration, negotiation can be supported using the claimant AI agent at an appropriate timing.

[0029] Furthermore, the negotiation support device of the sixth invention further comprises a prompt management unit that stores a respondent AI agent configuration prompt, which is a system prompt for configuring a respondent AI agent, which is an AI agent that supports negotiations on behalf of the respondent, in addition to any one of the first to third inventions; the agent means provides the respondent AI agent configuration prompt to the generating AI, provides the respondent AI agent with claim information, and performs respondent proxy processing, which is processing for interacting with the claimant, a third party, or other AI agents on behalf of the respondent; the respondent proxy processing is a negotiation support device that includes proxy processing, which includes receiving negotiation information from the claimant or a third party, or acquiring dialogue information or generated information output from an AI agent other than the respondent AI agent, passing the negotiation information or dialogue information or generated information to the respondent AI agent, obtaining a response from the respondent AI agent, and outputting negotiation information based on the response to the claimant, a third party, or an AI agent other than the respondent AI agent.

[0030] This configuration allows the respondent's AI agent to assist in negotiations.

[0031] Furthermore, the negotiation support device of the seventh invention, in contrast to the sixth invention, is a negotiation support device in which the agent means performs respondent proxy processing when the respondent agent cooperation conditions are met.

[0032] This configuration allows for the use of an AI agent representing the respondent to support negotiations at the appropriate time.

[0033] Furthermore, the negotiation support device of the eighth invention further comprises a prompt management unit that stores one third-party AI agent configuration prompt, which is a system prompt for configuring a third-party AI agent that is an AI agent that supports negotiations on behalf of a third party supporting negotiations, or two or more different third-party AI agent configuration prompts, with respect to any one of the first to third inventions, the agent means provides one or more different third-party AI agent configuration prompts to a generating AI, provides claim information to one or more third-party AI agents, and performs third-party proxy processing, which is processing for interacting with one or more of the claimant or respondent or other AI agents on behalf of one or more third parties, the third-party proxy processing is a negotiation support device that includes proxy processing that receives negotiation information from the claimant or respondent, or acquires dialogue information or generated information output from an AI agent other than the third-party AI agent, passes the negotiation information or dialogue information or generated information to the third-party AI agent, obtains a response from the third-party AI agent, and outputs negotiation information based on the response to an AI agent other than the claimant or respondent or the third-party AI agent.

[0034] This configuration allows a third-party AI agent to assist in negotiations.

[0035] Furthermore, the negotiation support device of this ninth invention, in contrast to the eighth invention, is a negotiation support device in which the agent means performs third-party proxy processing when the conditions for cooperation with a third-party agent are met.

[0036] This configuration allows for the use of a third-party AI agent to assist in negotiations at the appropriate time.

[0037] Furthermore, the negotiation support device of the tenth invention is a negotiation support device in which, with respect to the eighth or ninth invention, one or more different third-party AI agents are one or more AI agents selected from among a mediator AI agent who acts on behalf of a mediator, an arbitrator AI agent who acts on behalf of an arbitrator, a supporter AI agent who acts on behalf of a supporter, or a secretariat AI agent who acts on behalf of a secretariat.

[0038] This configuration allows for negotiation support through an appropriate third-party AI agent.

[0039] Furthermore, the negotiation support device of the eleventh invention further comprises a prompt management unit that stores a petitioner support AI agent configuration prompt, which is a system prompt for configuring a petitioner support AI agent, which is an AI agent that supports the petitioner, with respect to any one of the first to third inventions, and the agent means is a negotiation support device that performs petitioner support processing, which provides the petitioner support AI agent configuration prompt to a generating AI, provides petition information to the petitioner support AI agent, obtains a response from the petitioner support AI agent, and outputs the response to the petitioner.

[0040] With this configuration, the petitioner can be assisted by the petitioner support AI agent.

[0041] Furthermore, the negotiation support device of the twelfth invention, compared to the eleventh invention, is a negotiation support device in which the agent means performs petitioner support processing when the petitioner support agent cooperation conditions are met.

[0042] This configuration allows for the use of an AI agent to support petitioners at the appropriate time.

[0043] Furthermore, the negotiation support device of the thirteenth invention further comprises a prompt management unit that stores a respondent support AI agent configuration prompt, which is a system prompt for configuring a respondent support AI agent, which is an AI agent that supports the respondent, in addition to any one of the first to third inventions, and the agent means is a negotiation support device that provides the respondent support AI agent configuration prompt to a generating AI, provides claim information to the respondent support AI agent, obtains a response from the respondent support AI agent, and performs respondent support processing to output the response to the respondent.

[0044] With this configuration, the respondent can be supported by the respondent support AI agent.

[0045] Furthermore, the negotiation support device of the fourteenth invention, in contrast to the thirteenth invention, is a negotiation support device in which the agent means performs respondent support processing when the respondent support agent cooperation conditions are met.

[0046] This configuration allows for the use of an AI agent to support the respondent at the appropriate time.

[0047] Furthermore, the negotiation support device of the fifteenth invention further comprises a prompt management unit which stores a third-party support AI agent configuration prompt, which is a system prompt for configuring a third-party support AI agent, which is an AI agent that supports a third party assisting in negotiations, with respect to any one of the first to third inventions, and the agent means is a negotiation support device that provides the third-party support AI agent configuration prompt to a generating AI, provides claim information to the third-party support AI agent, obtains a response from the third-party support AI agent, and performs third-party support processing which outputs the response to the third party.

[0048] With this configuration, a third-party AI agent can assist a third party assisting in negotiations.

[0049] Furthermore, the negotiation support device of the sixteenth invention, compared to the fifteenth invention, is a negotiation support device in which the agent means performs third-party support processing when the conditions for cooperation with a third-party support agent are met.

[0050] This configuration allows for the use of a third-party AI agent to provide support to a third party at the appropriate time.

[0051] Furthermore, the negotiation support device of the seventeenth invention is a negotiation support device in which, with respect to the fifteenth or sixteenth invention, one or more different third-party A support I agents are one or more AI agents from among a mediator support AI agent, an arbitrator support AI agent, an arbitrator support AI agent, an aid provider support AI agent, or an office support AI agent.

[0052] This configuration allows for the support of third parties assisting in negotiations through appropriate third-party AI agents.

[0053] Furthermore, the negotiation support device of the eighteenth invention further comprises a prompt management unit which stores a neutrality judgment AI agent configuration prompt, which is a system prompt for configuring a neutrality judgment AI agent that performs a neutrality evaluation regarding negotiations with respect to any one of the first to seventeenth inventions, acquires a neutrality evaluation result which is the result of the evaluation of neutrality, and outputs it. The agent means provides the neutrality judgment AI agent configuration prompt to a generating AI, provides the neutrality judgment AI agent with one or two types of information from claim information and generated information which is negotiation information input by a third party or information output by a third-party AI agent, acquires the neutrality evaluation result from the neutrality judgment AI agent, and provides the neutrality evaluation means which outputs the neutrality evaluation result or information based on the neutrality evaluation result.

[0054] This structure allows for an assessment of neutrality in negotiations.

[0055] Furthermore, the negotiation support device of the nineteenth invention, in contrast to the eighteenth invention, is a negotiation support device in which the neutrality evaluation means outputs a neutrality evaluation result or information based on the neutrality evaluation result when the neutrality output conditions are met.

[0056] This configuration allows for the output of neutrality evaluation results at the appropriate time.

[0057] Furthermore, the negotiation support device of this 20th invention, in relation to any one of the first to nineteen inventions, uses the agent means to provide, using claim information, negotiation information received during negotiations, dialogue information indicating the content of conversations between the participants or the AI ​​agent and the AI ​​agent during negotiations, or generated information output by the AI ​​agent during negotiations, to provide, laws, articles, amendment history, court precedents, case commentaries, legal literature, papers, academic theories, forms, procedural information, industry information, industry-specific legal risks, industry practices, administrative documents, comparative legal information, foreign legal systems, or legal terms. This negotiation support device retrieves one or more pieces of information from a storage unit that contains either a collection of static information (like a dictionary), or a collection of dynamic information (such as application information, consultation details, party input, chat history, audio transcripts, phone logs, submitted documents, evidence, party comments, mediator comments, secretariat comments, AI proposal history, mediator selection / modification / rejection logs, settlement proposals, settlement / rejection results, performance status, recovery results, participation rate, response rate, and time to resolution), and passes that one or more pieces of information to a role-playing AI agent or support AI agent.

[0058] This configuration allows the AI ​​agent to learn appropriate information, thereby enabling the acquisition of appropriate information from the AI ​​agent. The negotiation support device of the present invention comprises a claim acquisition unit that acquires claim information having claim content information relating to the content of the claim, a claim storage unit that stores the claim information, and a negotiation support unit that supports negotiations between the claimant who made the claim and the respondent against whom the claim was made, and the negotiation support unit comprises an agent means that performs dialogue processing, which is the process of communicating with the respondent on behalf of the claimant, in at least a part of the phases from the start to the end of negotiations between the claimant and the respondent.

[0059] This configuration allows for the appropriate use of an AI agent in a negotiation support device that assists in negotiations between the petitioner and the respondent regarding a petition.

[0060] Furthermore, the negotiation support device of the present invention further comprises a condition management unit that stores, in association with one or more device linkage conditions which are conditions for an agent means to perform dialogue processing, and a processing identifier which identifies the processing to be executed, and the agent means is a negotiation support device that, in dialogue with the respondent, acquires the respondent's statements and, if the statements satisfy the device linkage conditions, executes the processing identified by the processing identifier associated with the device linkage conditions.

[0061] With this configuration, the negotiation support device can perform appropriate processing at the appropriate time when the respondent's statements meet the device linkage conditions.

[0062] Furthermore, the negotiation support device of the present invention, in addition to the above invention, has the following conditions for device linkage: the respondent's statement includes information indicating that negotiations will be conducted using the negotiation support device; it further comprises a negotiation receiving unit that receives negotiation information from the petitioner and the respondent; and a negotiation storage unit that stores negotiation information. The device linkage means is a negotiation support device that, when the device linkage conditions are met, transmits information for inputting negotiation information to the respondent's terminal.

[0063] With this configuration, if the respondent indicates to the AI ​​agent that they wish to conduct negotiations using a negotiation support device, the process can transition to negotiations using the negotiation support device.

[0064] Furthermore, the negotiation support device of the present invention further comprises, in addition to the above invention, a receiving unit that receives applicant input, which is the applicant's input, from the applicant's terminal, and a determination unit that determines whether or not the agent cooperation conditions, which are the conditions for performing dialogue processing by the agent means, are met based on the applicant input or the performance status information of the agreement content corresponding to the application received by the receiving unit, and the agent means is a negotiation support device that performs dialogue processing, which is the process for communicating with the respondent, when the determination unit determines that the agent cooperation conditions are met.

[0065] With this configuration, negotiations with the respondent using an AI agent can be conducted at the appropriate time, when the applicant's input satisfies the agent collaboration conditions.

[0066] Furthermore, in the negotiation support device of the present invention, the agent linkage condition is that the applicant input, which is a company notification, has been received, and the receiving unit is a negotiation support device that receives company notifications.

[0067] This configuration allows for negotiations with the respondent using an AI agent at an appropriate time after the petitioner has entered their petition information.

[0068] Furthermore, the negotiation support device of the present invention, in contrast to the above invention, has agent cooperation conditions such as the claim acquisition unit acquiring claimant input, which is new claim information, the receiving unit receiving claim information, and the claim acquisition unit acquiring the claim information received by the receiving unit.

[0069] This configuration allows for negotiations with the respondent using an AI agent at an appropriate time after the petitioner has entered their petition information.

[0070] Furthermore, the negotiation support device of the present invention, in relation to the above invention, has agent linkage conditions such that the performance status information of the agreement corresponding to the claim is information indicating that the agreement has not been fulfilled, and the judgment unit is a negotiation support device that determines whether or not the performance status information of the agreement corresponding to the claim satisfies the agent linkage conditions.

[0071] This configuration allows for negotiations with the respondent using an AI agent at an appropriate time after the petitioner has entered their petition information.

[0072] Furthermore, the negotiation support device of the present invention further comprises a claim management unit that stores one or more claim status information, including respondent information, including the respondent's telephone number, and the agent means is a negotiation support device that uses the respondent's telephone number to make a call to the respondent and engages in a telephone conversation with the respondent.

[0073] This configuration allows the AI ​​agent and the respondent to negotiate by telephone.

[0074] Furthermore, the negotiation support device of the present invention, in contrast to the above invention, has respondent information in the claim management department that includes the respondent's telephone number or other contact information, and the agent means only makes a call to the respondent and engages in telephone conversation with the respondent if the respondent information includes a telephone number.

[0075] This configuration allows the AI ​​agent and the respondent to negotiate by telephone.

[0076] Furthermore, the negotiation support device of the present invention, in addition to the above invention, is a negotiation support device in which the agent means makes a call to the respondent using the respondent's telephone number, but if the respondent does not answer the phone, it makes another call to the respondent at a predetermined timing.

[0077] With this configuration, if the respondent does not answer the call from the AI ​​agent, the AI ​​agent can call the respondent at the following time:

[0078] Furthermore, the negotiation support device of the present invention, in relation to the above invention, comprises an agent means which is a learning means that causes a generating AI to learn a manual or Q&A information relating to the negotiation support device, and an answering means that receives input from the petitioner or respondent, and if the input is a question relating to the negotiation support device, obtains an answer to the question from the generating AI and outputs the answer to the petitioner or respondent.

[0079] This configuration allows for appropriate responses to questions from the petitioner or respondent regarding the negotiation support device. [Effects of the Invention]

[0080] The negotiation support device according to the present invention can construct an appropriate AI agent to support negotiations between the claimant and the respondent.

[0081] Furthermore, the effects of the negotiation support device described above are not necessarily limited, and any of the effects described herein, or other effects that can be inferred from this specification, may be achieved in conjunction with or in lieu of the effects described above. [Brief explanation of the drawing]

[0082] [Figure 1] Conceptual diagram of information system A in Embodiment 1 [Figure 2] Block diagram of information system A [Figure 3] Block diagram of the negotiation support device 1 [Figure 4] A flowchart illustrating an example of the operation of the negotiation support device 1. [Figure 5] A flowchart illustrating an example of the operation of the negotiation support device 1. [Figure 6] A flowchart illustrating an example of the operation of the negotiation support device 1. [Figure 7] A flowchart illustrating an example of the same decision-making process. [Figure 8] A flowchart illustrating an example of the same dialogue processing. [Figure 9] A flowchart illustrating an example of the device linkage process. [Figure 10] A flowchart illustrating an example of the notification information configuration process. [Figure 11] A flowchart illustrating an example of the proposed content structure processing. [Figure 12] A flowchart illustrating an example of the agreement document generation process. [Figure 13] A flowchart illustrating an example of the output information configuration process. [Figure 14] A flowchart illustrating an example of operation for the petitioner's terminal 2. [Figure 15] A flowchart illustrating an example of operation of the respondent's terminal 3. [Figure 16] This diagram shows an example of the phase flow for processing the application. [Figure 17] A diagram showing an example of the petitioner management sheet. [Figure 18] A diagram showing an example of the application management form. [Figure 19] A diagram showing an example of the proposed content template. [Figure 20] A diagram showing an example of the agreement document template. [Figure 21] A diagram showing an example of the prompt. [Figure 22] A diagram showing an example of the prompt. [Figure 23] Figure showing an example of the same screen. [Figure 24] Figure showing an example of the same screen. [Figure 25] Figure showing an example of the same screen. [Figure 26] Figure showing an example of the same screen. [Figure 27] Figure showing an example of the same screen. [Figure 28] Figure showing an example of the same screen. [Figure 29] A diagram showing an example of the agent collaboration conditions management table. [Figure 30] This diagram shows an example of a device linkage conditions management table. [Figure 31] Figure showing an example of the same screen. [Figure 32] A diagram showing an example of the notification information. [Figure 33] Conceptual diagram of information system B in Embodiment 2 [Figure 34] Block diagram of the same information system B [Figure 35] Block diagram of the negotiation support device 5 [Figure 36] A flowchart illustrating an example of the operation of the negotiation support device 5. [Figure 37] A flowchart illustrating an example of the operation of the negotiation support device 5. [Figure 38] A flowchart illustrating an example of the operation of the negotiation support device 5. [Figure 39] A flowchart illustrating an example of the agent decision-making process. [Figure 40] A flowchart illustrating an example of processing on behalf of the petitioner. [Figure 41] A flowchart illustrating an example of the support process for the petitioner. [Figure 42] A flowchart illustrating an example of the process of handling the claim on behalf of the respondent. [Figure 43] A flowchart illustrating an example of the support process for the respondent. [Figure 44] A flowchart illustrating an example of the process for receiving negotiation information. [Figure 45] A flowchart illustrating an example of third-party processing. [Figure 46] A flowchart illustrating an example of third-party support processing. [Figure 47] A flowchart illustrating an example of the AI ​​agent configuration process for the petitioner. [Figure 48] A flowchart illustrating an example of the settlement decision-making process. [Figure 49] A flowchart illustrating an example of the settlement proposal generation process. [Figure 50] A flowchart illustrating an example of the neutrality assessment process. [Figure 51] A diagram showing an example of the petitioner management sheet. [Figure 52] A diagram showing an example of the agent collaboration conditions management table. [Figure 53] A diagram showing an example of the AI ​​agent configuration prompt for the petitioner. [Figure 54] This diagram shows an example of a configuration prompt for the AI ​​agent supporting the petitioner. [Figure 55] A diagram showing an example of the AI ​​agent configuration prompt for the respondent. [Figure 56] This diagram shows an example of a configuration prompt for the AI ​​agent supporting the respondent. [Figure 57] This diagram shows an example of a third-party AI agent configuration prompt. [Figure 58] This diagram shows an example of a configuration prompt for the third-party AI agent support. [Figure 59] This diagram shows an example of a neutral judgment AI agent configuration prompt. [Figure 60] This diagram shows an example of a settlement proposal generation AI agent configuration prompt. [Figure 61] Block diagram of the computer system in the above embodiment [Modes for carrying out the invention]

[0083] The following describes embodiments of the negotiation support device and the like with reference to the drawings. Note that components denoted by the same reference numerals in the embodiments perform similar operations, and therefore, further explanation may be omitted.

[0084] (Embodiment 1) In this embodiment, a negotiation support device is described in which an AI agent (hereinafter referred to as "the agent" as appropriate) that interacts with the respondent during the negotiation process between the petitioner and the respondent functions. The agent may make calls to the respondent or the petitioner, or may receive calls from the respondent or the petitioner.

[0085] The petitioner's petition against the respondent, in this context, refers to, for example, the settlement of a claim or the resolution of various disputes. The petition is usually a petition against the respondent. The respondent is the subject of the petition. The petition can be anything that is the subject of negotiation. Various disputes include, for example, disputes regarding the eviction of a building, disputes regarding an increase in building rent, negotiations regarding retirement benefits, and disputes regarding retirement conditions. A dispute can also be called a dispute or a disagreement. A claim is a right of the petitioner to demand a certain action from the debtor. The certain action here is usually the payment of money, but it can also be the delivery of goods, the provision of services, etc., and the action demanded by the claim is not limited. In other words, while it is preferable for the claim dealt with here to be a monetary claim, it does not matter whether it is a real claim or an action claim.

[0086] In this embodiment, a negotiation support device is described in which an agent performs dialogue processing, which is the process of communicating with the respondent on behalf of the applicant, during at least some phases of negotiations between the applicant and the respondent, from the start of negotiations to the end of negotiations. In this embodiment, a negotiation support device is described in which an agent performs dialogue processing during at least some phases of negotiations between the applicant and the respondent, from the start of negotiations to the completion of the fulfillment of the agreed-upon terms.

[0087] Here, "phase" refers to a stage in the process leading up to the completion of actions between the petitioner and the respondent in response to a claim. A phase can also be called a stage, phase, process, or step. The phase in which negotiations begin may be the company's own claim, as described later, or the claim, etc. The dialogue here is usually a real-time dialogue.

[0088] In this embodiment, we will describe a negotiation support device that performs processing for the negotiation support device, and which executes processing corresponding to the predetermined device linkage conditions, when those conditions are met.

[0089] In this embodiment, a negotiation support device including an agent that communicates with the respondent, for example, by telephone or chat, when predetermined agent collaboration conditions are met will be described.

[0090] In this embodiment, if the respondent does not answer the phone, the following negotiation support device will be described that will call the respondent at the following timings.

[0091] In this embodiment, a negotiation support device is described that obtains and outputs responses from an agent trained with response information in response to inquiries from the respondent or the petitioner.

[0092] In this specification, information X being associated with information Y means that information Y can be obtained from information X, or information X can be obtained from information Y, and the method of association is irrelevant. Information X and information Y may be linked, may reside in the same buffer, may information X be contained in information Y, or information Y may be contained in information X, and so on.

[0093] Furthermore, in this specification, selecting or determining information Z means obtaining information Z, obtaining a pointer to information Z, obtaining the ID of information Z, setting a flag on information Z, etc., and it is sufficient to be able to access information Z.

[0094] Figure 1 is a conceptual diagram of information system A in this embodiment. Information system A comprises a negotiation support device 1, one or more petitioner terminals 2, one or more respondent terminals 3, and one or more generation AI devices 4.

[0095] Negotiation support device 1 is a device that enables an agent to interact with the respondent during the negotiation process between the claimant and the respondent. Negotiation support device 1 may or may not include an agent. Negotiation support device 1 may be, for example, a cloud server or an ASP server, but the type is not limited.

[0096] Petitioner's Terminal 2 is the terminal used by the petitioner. Respondent's Terminal 3 is the terminal used by the respondent. Petitioner's Terminal 2 and Respondent's Terminal 3 are, for example, so-called personal computers, smartphones, or tablet devices, but the type is not specified.

[0097] The Generating AI Device 4 is a device that has the functionality of a Generating AI. Here, the Generating AI Device 4 typically has the functionality of a text generation AI. The Generating AI may be, for example, ChatGPT (registered trademark) or Gemini, but is not limited to that. The Generating AI Device 4 may be, for example, a cloud server or an ASP server, but is not limited to that type. The Generating AI Device 4 will be referred to as the Generating AI as appropriate. The Negotiation Support Device 1 may also have the functionality of a Generating AI. In this case, the Generating AI Device 4 is not necessary. The Negotiation Support Device 1 does not need to use the functionality of a Generating AI. In this case as well, the Generating AI Device 4 is not necessary. The Generating AI may also be referred to as LLM, etc.

[0098] Negotiation support device 1 and one or more petitioner terminals 2, negotiation support device 1 and one or more respondent terminals 3, and negotiation support device 1 and one or more generating AI devices 4 can communicate with each other via a network such as the Internet.

[0099] The negotiation support device 1 comprises a storage unit 11, a receiving unit 12, a processing unit 13, and a transmission unit 14. The storage unit 11 comprises a claim management unit 111, a condition management unit 112, and a response information storage unit 113. The receiving unit 12 comprises a claim acquisition unit 121, a confirmation receiving unit 122, a negotiation receiving unit 123, and an instruction receiving unit 124. The processing unit 13 comprises a claim storage unit 131, an update unit 132, a confirmation storage unit 133, a negotiation support unit 134, a negotiation storage unit 135, a proposal content acquisition unit 136, a judgment unit 137, an agreement document generation unit 138, and an information acquisition unit 139. The negotiation support unit 134 comprises an agent means 1341. The agent means 1341 comprises a device linkage means 13411, a learning means 13412, and a response means 13413. The transmission unit 14 includes a petition notification unit 141, a screen transmission unit 142, a proposal content transmission unit 143, a negotiation information output unit 144, a necessary information transmission unit 145, an agreement document output unit 146, and an information transmission unit 147.

[0100] The petitioner terminal 2 comprises a petitioner storage unit 21, a petitioner reception unit 22, a petitioner processing unit 23, a petitioner transmission unit 24, a petitioner reception unit 25, and a petitioner output unit 26.

[0101] The respondent terminal 3 comprises a respondent storage unit 31, a respondent reception unit 32, a respondent processing unit 33, a respondent transmission unit 34, a respondent receiving unit 35, and a respondent output unit 36.

[0102] The storage unit 11, which constitutes the negotiation support device 1, stores various types of information. These types of information include, for example, application status information, various screen information, various conditions, response information, and various templates, which will be described later.

[0103] Screen information refers to the information used to construct a screen. Screen information can be written in HTML, XML, etc., but the method of implementation is not limited. Screen information is associated with, for example, selection objects within other screen information. A selection object is an object that can be selected. Examples of selection objects include buttons, menu items, or checkboxes.

[0104] The various templates include, for example, notification information templates, proposal content templates, and agreement document templates. A notification information template is a template document for composing notification information, as described later, and is a document with variables. A proposal content template is a template document for composing proposal content, and is a document with variables. An agreement document template is a template document for composing an agreement document, and is a document with variables.

[0105] The petition management unit 111 stores one or more petition status information. The petition status information preferably includes respondent information, which is described later. The respondent information preferably includes the respondent's telephone number.

[0106] Application status information refers to information related to an application. For example, application status information is information that can identify the current phase of one of two or more phases from application to agreement. The agreement here refers to an agreement between the applicant and the respondent regarding the handling of the application. Application status information can be said to be information for managing the status of an application. Application status information is information that corresponds to an application identifier. An application identifier is information that identifies an application. An application identifier is, for example, the application ID or the name of the application (for example, the name of the application case). Application status information includes, for example, applicant information, respondent information, application content information, confirmation information, negotiation information, proposed content, agreement documents, phase identification information, and various time information. The applicant is, for example, the applicant. The respondent is, for example, the debtor. Application status information may also be associated with, for example, one or more pieces of pre-application information. Pre-application information is information prior to the application. Pre-application information is, for example, company notices.

[0107] The application status information may include information on the status of implementation of the agreement. Information on the status of implementation of the agreement refers to information that identifies the status of implementation of the agreement. For example, this information may include information indicating that the agreement has been implemented, information indicating that the agreement has not been implemented, the amount paid, the date of payment completion, and the date of eviction. The process by which the information on the status of implementation of the agreement is accumulated in the application status information is not specified. For example, the information on the status of implementation of the agreement may be information entered by the applicant or respondent, or it may be information that is automatically acquired.

[0108] Self-notification refers to a notification from the applicant to the respondent prior to filing a claim using Negotiation Support Device 1. Notifications from the applicant to the respondent include notifications made by an agent, such as a lawyer. Self-notification usually consists of information about the applicant's own claim. An applicant's claim is when the applicant notifies (files a claim) the respondent themselves. Self-notification includes, for example, information from emails or certified mail sent by the applicant to the respondent requesting them to complete their response to the claim. Note that "prior to filing a claim" means before sending the claim information to Negotiation Support Device 1. Self-notification corresponds to, for example, the respondent's contact information. Self-notification includes, for example, invoices and / or information related to invoices sent by the applicant to the respondent, and self-claims and / or information related to self-claims sent by the applicant to the respondent. Information related to invoices includes, for example, the invoice issuance date and payment due date. Information related to self-claims includes, for example, the date the self-claim was filed. Furthermore, the process by which pre-filing information is accumulated in the petition management department 111 is not relevant.

[0109] Application status information is, for example, debt status information. Debt status information can be said to be information related to debt processing. Debt status information can be said to be, for example, information for managing the status of debts. Debt status information is information that corresponds to the application identifier.

[0110] Petitioner information refers to information about the petitioner. Petitioner information includes, for example, a petitioner identifier and one or more petitioner attribute values. The petitioner identifier is information that identifies the petitioner. The petitioner identifier includes, for example, the petitioner's ID, the petitioner's name, the petitioner's telephone number, the petitioner's email address, and the IP address of petitioner terminal 2. The petitioner attribute values ​​include, for example, the petitioner's payment destination information, the petitioner's company name, the name of the representative of the company that is the petitioner, and the petitioner type indicating whether the petitioner is a corporation or an individual. Payment destination information is information that the respondent uses to make payments to the petitioner. Payment destination information includes, for example, bank account information, credit card information, and online payment information. Bank account information is information about the bank account necessary for bank transfers and includes, for example, the bank name, branch name, and bank account number. Credit card information is information necessary for making payments using a credit card and includes, for example, the credit card number and expiration date. Online payment information is information necessary for making payments using online payment and includes, for example, an ID for online payment. Note that bank account information can be referred to as account information as appropriate.

[0111] Respondent information refers to information about the respondent. Respondent information includes, for example, a respondent identifier and one or more respondent attribute values. The respondent identifier is information that identifies the respondent. The respondent identifier may include, for example, the respondent's ID, the respondent's name, the respondent's telephone number, the respondent's email address, and the IP address of respondent terminal 3. Respondent attribute values ​​may include, for example, the respondent's contact information, the respondent's company name, the name of the representative of the respondent company, the respondent's address, and the respondent type indicating whether the respondent is a corporation or an individual. Contact information may include, for example, an email address, a telephone number, and an ID for social media (e.g., Line®, Slack®).

[0112] Claim information refers to information that identifies the content of a claim. This information may include, for example, the name of the claim, the claimed amount, and the original payment due date. If the claim is a right to demand the return of borrowed items, the claim information may include, for example, information about the items to be returned and information identifying the return address.

[0113] Confirmation information is information indicating that the respondent has confirmed the petitioner's petition. Confirmation by the respondent means that the respondent acknowledges that the petition is their case. Confirmation information is, for example, a flag, but it may be replaced by information that should be stored after the confirmation phase (e.g., negotiation information) being contained within the petition status information. Confirmation information is, for example, confirmation time information. Confirmation time information is information that identifies when the respondent confirmed the petition. Confirmation time information is, for example, the date and time the information was received from respondent terminal 3. The time information specifying the time may be just the date, or the date and time, etc. The granularity of the time information is not specified. Confirmation information only needs to be information that allows the respondent to understand that the petition has been confirmed.

[0114] Negotiation information refers to information relating to negotiations between the petitioner and the respondent regarding the processing of the petition. Negotiation information can be said to be information that identifies the content of the negotiations between the petitioner and the respondent regarding the processing of the petition. The information that identifies the content of the negotiation is the content of the negotiation. Negotiation information may include an inputter identifier. An inputter identifier is information that identifies the person who proposed the content of the negotiation. An inputter identifier may be, for example, information that identifies the petitioner or information that identifies the respondent.

[0115] The negotiations here concern the processing of the application. The negotiations are between the applicant and the respondent until an agreement is reached regarding the processing of the application. Negotiation information includes, for example, an application processing identifier. An application processing identifier is information that identifies the method for processing the application. An application processing identifier may include, for example, an ID and the name of the application processing (e.g., "Pay now by credit card," "Pay in full by the end of next month," "Pay in installments at the end of each month," "Proposal to revise the amount," "I have no recollection of the claim," "Already paid directly," "Request online mediation"). An application processing identifier may also include, for example, a payment method identifier that identifies the means of payment (e.g., "credit card," "bank transfer," etc.).

[0116] The proposed content is a document that describes matters agreed upon by both parties when processing the application, which is identified by the application processing identifier. In other words, the proposed content is information that indicates a proposal regarding the processing of the application. The proposed content includes information about the applicant (e.g., the applicant's name) and the respondent (e.g., the respondent's name), and also includes information about the negotiation (e.g., the monthly amount in installment payments).

[0117] An agreement document is a document created when negotiations regarding the handling of a claim reach an agreement, and it contains the content of those negotiations. The agreement document may include, for example, information about the claimant (e.g., the claimant's name), information about the respondent (e.g., the respondent's name), the method of payment by the respondent to the claimant, and the amount to be paid. The agreement document may also be identical to the final proposal made during negotiations.

[0118] Phase identification information refers to information that identifies the current phase among two or more phases from the time the application is filed until the applicant and the respondent agree on how to process the application. The phases that can be identified by the phase identification information do not have to be all of the two or more phases from the time the application is filed until the applicant and the respondent agree on how to process the application. Phase identification information is, for example, a phase identifier, but it does not have to be explicitly present in the application status information. Phase identification information only needs to be information that allows the phase to be determined from the application status information. Phase identification information may be information that can be determined from information that the application status information has or does not have. For example, if the application status information has information at the time of application and at the time of confirmation, but does not have negotiation information, the current phase can be determined to be the "negotiation" phase. In such a case, the phase identification information is a phase identifier determined from the time of application and at the time of confirmation, or from the time of application and at the time of confirmation and negotiation information that does not exist.

[0119] The two or more phases from the filing of a claim until the claimant and respondent agree on how to handle the claim are, for example, "claim filed," "awaiting confirmation," "confirmed," "negotiating," and "agreement reached." The "claim filed" phase is the phase in which the claimant files a claim with the respondent. The "awaiting confirmation" phase is the phase in which the claimant is waiting for confirmation from the respondent regarding the filing of the claim. Note that the "awaiting confirmation" phase may be the same as the "claim filed" phase. The "confirmed" phase is the phase in which the respondent has confirmed the filed claim. The "negotiating" phase is the phase in which the claimant and respondent are negotiating how to handle the claim. The "agreement reached" phase is the phase in which negotiations for the handling of the claim have reached an agreement. Note that the two or more phases from the filing of a claim until the claimant and respondent agree on how to handle the claim do not necessarily have to include the "claim filed" phase and the "agreement" phase.

[0120] Furthermore, the phases handled by Negotiation Support Device 1 may include phases after the negotiation agreement has been reached. Such phases may include, for example, "Application in Progress" and "Application Processing Completed." The "Application in Progress" phase is the phase after the negotiation agreement has been reached and the application is being processed (for example, during payment). The "Application Processing Completed" phase is the phase when the application processing has been completed. Ideally, the phases handled by Negotiation Support Device 1 should include "Application Filed," "Awaiting Confirmation," "Confirmed," "Negotiating," "Agreement Reached," "Application in Progress," and "Application Processing Completed," but it may also include some of these phases. In addition, the phases handled by Negotiation Support Device 1 do not necessarily have to include all phases from the filing of the application to the agreement between the applicant and the respondent regarding the processing of the application.

[0121] Furthermore, various types of time information include, for example, application information, confirmation information, negotiation information, agreement information, and application processing completion information. Application information is information indicating when an application was filed. For example, application information is information indicating when the application acquisition unit 121 received the application information. Negotiation information is information indicating when negotiations took place. For example, negotiation information is information indicating when the negotiation receiving unit 123 received the negotiation information. Agreement information is information indicating when an agreement was reached. For example, agreement information is information indicating when the judgment unit 137 determined that the agreement conditions described later had been met. Application processing completion information is information indicating when the processing of the application was completed.

[0122] The condition management unit 112 stores one or more conditions. These conditions may include, for example, agent collaboration conditions, device collaboration conditions, proposal content creation conditions, or agreement conditions. It goes without saying that these conditions may also be embedded within the program.

[0123] Each of the one or more device linkage conditions in the condition management unit 112 is associated with, for example, a processing identifier. Each of the one or more device linkage conditions is also associated with, for example, a phase identifier. The processing identifier here is information that identifies the process to be executed. The processing identifier is, for example, a function name, module name, method name, the program itself, or a link to the program. The program itself is a function, module, method, etc. It is sufficient that the process is implemented using the processing identifier. The processes identified by the processing identifier are, for example, a process to send the respondent's input to the petitioner, a process to send the petitioner's input to the respondent, a process to log in in order to use the negotiation support device 1, a process to assist in logging in in order to use the negotiation support device 1, and a process to obtain payment account information and send it to the respondent.

[0124] Agent linkage conditions are conditions for dialogue processing by the AI ​​agent. It can be said that agent linkage conditions are conditions for dialogue processing by agent means 1341. Agent linkage conditions are, for example, conditions based on the current phase. Agent linkage conditions are, for example, conditions based on the claimant's input. Agent linkage conditions are, for example, conditions based on the respondent's input. Agent linkage conditions are, for example, conditions based on the current phase and the claimant's input. Agent linkage conditions include, for example, conditions for the phase identifier and conditions for the claimant's input or respondent's input. Agent linkage conditions are, for example, conditions regarding information on the status of compliance with the agreement.

[0125] Agent integration conditions include, for example, the receipt of a company notification from the claimant. Agent integration conditions include, for example, the receipt of claim information from the claimant. Agent integration conditions include, for example, the information on the status of performance of the agreement is "information indicating that the agreement has not been performed." Claimant input refers to input from the claimant. Respondent input refers to input from the respondent. Furthermore, the functions of the AI ​​agent are, for example, functions realized by the generative AI.

[0126] Device linkage conditions are the conditions under which the negotiation support device 1 performs predetermined processing. Predetermined processing is, for example, the processing specified by the processing identifier. Predetermined processing is, for example, the processing specified by the processing identifier paired with the device linkage conditions. Device linkage conditions are, for example, conditions based on respondent input. Device linkage conditions are, for example, that during the dialogue between agent means 1341 and the respondent, the input from the respondent indicates predetermined matters. Input indicating predetermined matters includes, for example, "information indicating a desire to negotiate using negotiation support device 1," "information indicating a response to the claim by the end of next month," "information indicating a willingness to pay in installments," "information indicating a lack of awareness of the claim," and "information indicating a response to the claim has already been taken."

[0127] The conditions for creating a proposal are the conditions for creating a proposal. For example, the conditions for creating a proposal are that the negotiation information includes a claim processing identifier that indicates a specific method from two or more methods of claim processing (that a specific claim processing identifier has been selected). A specific claim processing identifier is, for example, from the claim processing identifiers "Pay now by credit card," "Pay in full by the end of next month," "Pay in installments at the end of each month," "Proposal to revise the amount," "I have no recollection of the claim," "Already paid directly," and "Want online mediation," the specific claim processing identifier is "Pay now by credit card," "Pay in full by the end of next month," "Pay in installments at the end of each month," or "Proposal to revise the amount." For example, the conditions for creating a proposal are that the negotiation information includes the installment amount. For example, the conditions for creating a proposal are that the received negotiation information includes information that has not been included in any previously received negotiation information.

[0128] Agreement conditions are the conditions used to determine whether an agreement has been reached regarding the processing of the application. Agreement conditions are, for example, conditions based on negotiation information. Agreement conditions are, for example, that the received negotiation information includes an application processing identifier indicating a specific method of processing the application from two or more methods (that a specific application processing identifier has been selected). A specific application processing identifier is, for example, from the debt processing identifiers "Pay now by credit card," "Pay in full by the end of next month," "Pay in installments at the end of each month," "Proposal to revise the amount," "I have no recollection of the claim," "Already paid directly," and "Want online mediation," the application processing identifiers are "Pay now by credit card" and "Pay in full by the end of next month." Agreement conditions are, for example, that the received negotiation information includes a payment method identifier. Agreement conditions are, for example, that negotiation information indicating that the respondent has agreed to the negotiation information sent by the applicant has been received. Agreement conditions are, for example, that negotiation information indicating that the applicant has agreed to the negotiation information sent by the respondent has been received.

[0129] The response information storage unit 113 stores response information that serves as the basis for answers to questions from the respondent or the petitioner. These questions are, for example, questions regarding the use of the negotiation support device 1. The response information consists of one or more types of information, such as the manual for the negotiation support device 1, the Q&A information provided by the negotiation support device 1, or the FAQ information.

[0130] The prompt management unit 114 stores various prompts. These prompts include, for example, negotiator prompts, inquiry prompts, agreement decision prompts, or inquiry decision prompts. Negotiator prompts include, for example, post-company notification prompts or post-claim prompts.

[0131] A negotiator prompt is a prompt used to configure an AI agent for negotiating with the respondent.

[0132] A post-company notification prompt is a prompt used to configure an AI agent to negotiate with the respondent after the company has issued a company notification. For example, a post-company notification prompt might look like this: "You are the agent who will negotiate with the respondent on behalf of the petitioner regarding the following notification from the petitioner to the respondent. [Notification]<Company Notification> Please call the respondent's contact number below to negotiate. [Contact Number]<Respondent.Contact Information>..." Note that strings enclosed in "<" and ">" in the prompt are variables.

[0133] A post-filing prompt is a prompt used to configure an AI agent to confirm the respondent's intentions regarding a petition. For example, a post-filing prompt might say, "You are the agent who will negotiate with the respondent on behalf of the petitioner regarding the following petition from the petitioner to the respondent. [Petition]<Petition Information> Please call the respondent's contact number below to negotiate. [Contact Information]<Respondent's Contact Information>..."

[0134] An inquiry prompt is a prompt used to obtain answers from the generating AI to questions regarding the use of Negotiation Support Device 1. For example, an inquiry prompt is used to prompt the AI ​​agent to answer questions regarding the use of Negotiation Support Device 1. An inquiry prompt may look like this: "The following question was asked. Please answer the following user question using the learned answer information (manual, Q&A collection). [User Question]<User Input>" The user input refers to either the petitioner's input or the respondent's input. The variable <User Input> here is usually assigned the user's response during the conversation. Preferably, the user's response assigned is a string. The string may also be information obtained by character recognition of the user's speech. The user is either the petitioner or the respondent.

[0135] An agreement decision prompt is a prompt used to determine whether or not the user has agreed to the claim based on user input. An agreement decision prompt might look like this: "Please determine whether or not you have agreed to the claim based on the following user input. Output '1' if you have agreed, and '0' if you have not. [User Input]<User Input>" The variable <User Input> here is usually assigned negotiation information.

[0136] An inquiry judgment prompt is a prompt used to determine whether a user's statement is an inquiry or not. An example of an inquiry judgment prompt is: "Is the following response a question to the negotiation support device? If it is a question, return "1"; if it is not a question, return "0". Also, if you are ending the conversation, return "-1". [Response]<Response>"

[0137] The receiving unit 12 receives applicant input from the applicant terminal 2. Applicant input includes, for example, company notices, application information, and negotiation information.

[0138] The receiving unit 12 receives the respondent's input from the respondent's terminal 3. The respondent's input may include, for example, confirmation information and negotiation information.

[0139] The application acquisition unit 121 acquires application information. For example, the application acquisition unit 121 receives application information from the applicant terminal 2. For example, the application acquisition unit 121 receives a case registration containing application information from the applicant terminal 2. The application acquisition unit 121 may also read application information from the storage unit 11 or an external device (not shown). The method by which the application acquisition unit 121 acquires application information is not limited.

[0140] Petition information refers to information relating to a petition. Petition information may also be considered to include information relating to claims, etc., that may be petitioned in the future, and the term "petition information" is interpreted broadly. Petition information may include, for example, a petitioner identifier that identifies the petitioner, a respondent identifier that identifies the respondent, and petition content information. Petition information may also include, for example, petitioner information that has a petitioner identifier, and respondent information that has a respondent identifier. Case registration is an instruction to register the case that is the subject of the petition.

[0141] The application acquisition unit 121 may, for example, acquire application time information from a clock (not shown), and constitute application information that includes said application time information and received application information, and which is stored information.

[0142] The confirmation receiving unit 122 receives confirmation information regarding the confirmation of the application from the respondent's terminal 3. The confirmation receiving unit 122 usually receives confirmation information from the respondent's terminal 3 in response to the notification information described later being sent to the respondent. The confirmation information is usually information indicating that the respondent has acknowledged the existence of the application (e.g., a claim). The confirmation information received by the confirmation receiving unit 122 is usually associated with the application identifier. The confirmation information received by the confirmation receiving unit 122 is, for example, a flag indicating that the application has been confirmed, or the respondent identifier.

[0143] The confirmation receiving unit 122 may receive non-existent information. Non-existent information is information indicating the respondent's claim that the matter related to the application (for example, a claim) does not exist.

[0144] The negotiation receiving unit 123 receives negotiation information from the petitioner or the respondent. For example, the negotiation receiving unit 123 receives negotiation information from the petitioner's terminal 2. For example, the negotiation receiving unit 123 receives negotiation information from the respondent's terminal 3.

[0145] The negotiation receiving unit 123 receives negotiation information from the petitioner's terminal 2 and the respondent's terminal 3, for example, after the confirmation receiving unit 122 has received confirmation information. The negotiation receiving unit 123 receives negotiation information alternately from the petitioner's terminal 2 and the respondent's terminal 3, for example. The negotiation is usually for processing the petition (for example, extinguishment of the claim). The negotiation information received by the negotiation receiving unit 123 is usually associated with the petition identifier. The negotiation information received by the negotiation receiving unit 123 is usually associated with the petitioner identifier of the petitioner who sent the negotiation information or the respondent identifier of the respondent who sent the negotiation information.

[0146] The negotiation receiving unit 123 receives negotiation information from the respondent terminal 3, for example, which includes a petition processing identifier that identifies one of two or more methods of petition processing. The negotiation receiving unit 123 also receives negotiation information from the petitioner terminal 2 or respondent terminal 3, which is the petitioner's proposal, including the starting month for installment payments for the petition processing and the payment amount for each month.

[0147] The negotiation receiving unit 123 receives negotiation information, for example, proposals regarding revisions to the payment amount, from the petitioner's terminal 2 or the respondent's terminal 3.

[0148] The negotiation receiving unit 123 receives negotiation information from the respondent terminal 3, including, for example, "information indicating that the claim is unfamiliar" and "information indicating that direct payment has already been made." The negotiation receiving unit 123 also receives negotiation information from the petitioner terminal 2 or respondent terminal 3, including, for example, "information indicating that online mediation is desired."

[0149] The instruction receiving unit 124 receives various instructions from the petitioner's terminal 2 or the respondent's terminal 3. These instructions include, for example, output instructions, proposed content instructions, and agreement document instructions.

[0150] An output instruction is an instruction to output application status information from the application management unit 111. An output instruction is associated with an application identifier. For example, an output instruction has an application identifier. An instruction to output application status information can be considered to include an instruction to output some of the information of the application status information. A proposal content instruction is an instruction to output the proposal content. An agreement document instruction is an instruction to output the agreement document.

[0151] The processing unit 13 performs various processes. These processes include, for example, those performed by the application storage unit 131, the update unit 132, the confirmation storage unit 133, the negotiation support unit 134, the negotiation storage unit 135, the proposal content acquisition unit 136, the judgment unit 137, the agreement document generation unit 138, or the information acquisition unit 139.

[0152] The application storage unit 131 stores the application information acquired by the application acquisition unit 121 in the application management unit 111, associating it with the application identifier. The stored application information constitutes the application status information of the application management unit 111.

[0153] The claim storage unit 131 generates a unique claim identifier when the information acquisition unit 139 receives claim information from the claimant terminal 2, and stores the claim information in the claim management unit 111 in association with the claim identifier. The method for generating the unique claim identifier is not limited. The claim storage unit 131 obtains a unique numerical value by adding 1 to the most recently generated claim identifier.

[0154] The update unit 132 is phase identification information that identifies the phase indicating "company notification" when the receiving unit 12 receives a company notification, and stores the phase identification information associated with the claim identifier in the claim management unit 111. When the receiving unit 12 receives a company notification, the update unit 132 may also obtain a unique claim identifier and store the phase identification information that identifies the phase indicating "company notification" associated with the claim identifier in the claim management unit 111.

[0155] The update unit 132 updates the phase identification information associated with the application identifier in response to the application notification unit 141 sending notification information to the respondent. The update unit 132 usually stores information in the application management unit 111 indicating that the current phase is phase "awaiting confirmation". The update unit 132 may store the notification information in the application management unit 111 associated with the application identifier, or it may store a flag indicating that the notification information has been sent in the application management unit 111 associated with the application identifier. The update unit 132 only needs to perform processing to identify that the phase has advanced as a result of the transmission of the notification information. Such processing may be replaced by the processing in which the application storage unit 131 stores application information. For example, the update unit 132 updates the phase identification information to phase identification information indicating that the current phase is phase "awaiting confirmation".

[0156] The update unit 132 updates the phase identification information associated with the application identifier in response to the confirmation receiving unit 122 receiving confirmation information. Here, the update unit 132 stores in the application management unit 111 information indicating that the current phase is phase "negotiating". The update unit 132 may store the confirmation information in the application management unit 111 associated with the application identifier, or it may store in the application management unit 111 a flag indicating that confirmation information has been received associated with the application identifier. The update unit 132 only needs to perform processing to identify that the phase has advanced due to the receipt of confirmation information. Such processing may be replaced by the processing in which the confirmation storage unit 133 stores the confirmation information. The update unit 132 updates the phase identification information to phase identification information indicating that the current phase is phase "negotiating".

[0157] The confirmation storage unit 133, for example, acquires confirmation information to be stored in response to the confirmation receiving unit 122 receiving confirmation information, associates the confirmation information with the application identifier, and stores it in the application management unit 111.

[0158] For example, when the agent means 1341 is in dialogue with the respondent, if the input (statement) from the respondent includes "information indicating that the application has been confirmed", the confirmation storage unit 133 acquires the confirmation information and stores it in the application management unit 111, associating it with the application identifier.

[0159] The confirmation storage unit 133 acquires confirmation time information, which is time information, from, for example, a clock (not shown), and stores the confirmation information, including the confirmation time information, in the application management unit 111, associating it with the application identifier. Note that the stored confirmation information may also be confirmation time information. In other words, the received confirmation information and the stored confirmation information may be different information. The confirmation information only needs to be information that allows the respondent to determine that they have confirmed the application.

[0160] Negotiation Support Unit 134 assists in negotiations between the petitioner and the respondent. For example, Negotiation Support Unit 134 assists in negotiations with the respondent using an AI agent, which is a virtual agent of the petitioner.

[0161] Agent means 1341 performs dialogue processing, which is the process of communicating with the respondent on behalf of the petitioner, for example, in at least some phases of negotiations between the petitioner and the respondent, from the start to the end of negotiations. Agent means 1341 performs dialogue processing with the respondent, for example, in at least some phases of negotiations between the petitioner and the respondent, from the start to the completion of the fulfillment of the agreement. The dialogue processing performed by agent means 1341 may also be performed by an AI agent. The dialogue processing performed by agent means 1341 may also be performed by having an AI agent perform dialogue processing.

[0162] Agent means 1341 provides the negotiator prompt from the prompt management unit 114 to the generating AI. Agent means 1341 also provides the respondent input, which is input from the respondent, to the generating AI. Agent means 1341 outputs the response from the generating AI to the respondent.

[0163] Agent means 1341, for example, provides the negotiator prompt from the prompt management unit 114 to the generating AI. Agent means 1341, for example, obtains the respondent's telephone number from the storage unit 11 and makes a call to the respondent using that telephone number. Agent means 1341 then outputs the response from the generating AI as voice to the respondent's telephone. Agent means 1341, for example, obtains input from the respondent via telephone and provides it to the generating AI. Here, agent means 1341 may, for example, perform speech recognition on the voice from the respondent via telephone, obtain a string, and provide the string to the generating AI, or it may provide the voice information from the respondent via telephone to the generating AI.

[0164] When the agent means 1341 meets the agent cooperation conditions, and the decision unit 137 determines this, it performs a dialogue process, which is a process for communicating with the respondent.

[0165] Agent method 1341 uses the respondent's telephone number to call the respondent and converse with the respondent by telephone.

[0166] Agent means 1341 may prefer to call the respondent and converse with the respondent by telephone only if the respondent information of the respondent in this application includes a telephone number. Agent means 1341 may prefer to determine whether the respondent information includes a telephone number, and only if it does, call the respondent and converse with the respondent by telephone.

[0167] Agent means 1341, using the respondent's telephone number, attempts to call the respondent, but if the respondent does not answer, it is preferable to call the respondent again at a predetermined time. This predetermined time may be, for example, after a predetermined period, after a predetermined time, or after a predetermined number of days.

[0168] Agent linkage conditions include, for example, the acquisition of a claim unit 121 acquiring new claim information, and the receiving unit 12 receiving a notification from the company. Claim information is also an example of claimant input.

[0169] The device linkage means 13411, in dialogue with the respondent, obtains the respondent's statements (respondent input), and if the statements satisfy the device linkage conditions, executes the process specified by the processing identifier associated with the device linkage conditions.

[0170] The device linkage means 13411 transmits information for inputting negotiation information to the respondent's terminal, for example, when the device linkage conditions are met.

[0171] The device integration condition is that the respondent's statement includes information indicating that negotiations will be conducted using negotiation support device 1.

[0172] The learning means 13412 causes the generating AI to learn the answer information from the answer information storage unit 113. In other words, the learning means 13412 provides the generating AI with the answer information from the answer information storage unit 113. For example, the learning means 13412 causes the generating AI to learn manuals or Q&A information related to the negotiation support device 1.

[0173] The response means 13413 receives input from the petitioner or respondent, and if the input is a question concerning the negotiation support device 1, it passes the question to the generating AI, obtains an answer to the question from the generating AI, and outputs the answer to the petitioner or respondent.

[0174] The negotiation storage unit 135 stores negotiation information. For example, the negotiation storage unit 135 stores the negotiation information received by the negotiation receiving unit 123 in the application management unit 111, associating it with an applicant identifier that identifies the applicant who sent the negotiation information, an applicant identifier that identifies the respondent who sent the negotiation information, and an application identifier. The negotiation storage unit 135 may also store all or part of the respondent input, which is input from the respondent acquired by the agent means 1341, as negotiation information.

[0175] The negotiation storage unit 135 may, upon acquiring negotiation information, acquire negotiation time information from a clock (not shown) and store the negotiation information including said negotiation time information. The negotiation storage unit 135 may, upon acquiring negotiation information, acquire an inputter identifier that identifies whether the person who entered the negotiation information is the petitioner or the respondent, and store the negotiation information including said inputter identifier.

[0176] The proposal content acquisition unit 136 acquires the proposal content using the acquired negotiation information if the acquired negotiation information matches the conditions for creating the proposal content. It is preferable for the proposal content acquisition unit 136 to acquire the proposal content using the application status information.

[0177] The proposal content acquisition unit 136 acquires information about the petitioner and information about the respondent from the petition management unit 111, and acquires proposal content, which is information that includes the information about the petitioner and the respondent and indicates a proposal corresponding to one method of processing the petition.

[0178] The proposal content acquisition unit 136 constructs the proposal content by, for example, substituting the applicant's information and the respondent's information into a proposal content template paired with one or more application processing identifiers. The proposal content acquisition unit 136 constructs the proposal content by, for example, substituting the applicant's information (e.g., applicant's name), the respondent's information (e.g., respondent's name), and some or all of the negotiation information (e.g., payment amount) into a proposal content template paired with one or more application processing identifiers.

[0179] The determination unit 137 determines whether the agent linkage conditions, which are the conditions for dialogue processing by the agent means 1341, are met, based on the applicant input or the performance status information of the agreement content corresponding to the application received by the receiving unit 12. For example, the determination unit 137 determines whether the performance status information of the agreement content corresponding to the application of interest satisfies the agent linkage conditions. The timing of the determination unit 137's determination of whether the performance status information satisfies the agent linkage conditions is irrelevant. For example, the determination unit 137 determines whether the performance status information satisfies the agent linkage conditions periodically (for example, once a day at 12:00). For example, the determination unit 137 determines whether the performance status information satisfies the agent linkage conditions upon receiving instructions from the applicant.

[0180] The decision unit 137 uses the acquired negotiation information to determine whether the agreement conditions have been met. For example, the decision unit 137 determines whether the acquired negotiation information has been met. For example, the decision unit 137 determines whether the numerical values ​​(e.g., amount, period, timing) in the received negotiation information have been met. For example, the decision unit 137 uses natural language processing on the received negotiation information to obtain information on whether the user agrees or not. For example, the decision unit 137 provides the agreement decision prompt, the acquired negotiation information, and the agreement conditions to the generating AI and obtains information from the generating AI indicating whether the agreement conditions have been met or not.

[0181] The agreement document generation unit 138 generates an agreement document using one or more pieces of negotiation information if the acquired negotiation information satisfies the agreement conditions. It is preferable for the agreement document generation unit 138 to store the generated agreement document in the application management unit 111, paired with the application identifier.

[0182] The agreement document generation unit 138 generates an agreement document by, for example, substituting the applicant's information, the respondent's information, and information contained in one or more negotiation documents into the agreement document template in the storage unit 11.

[0183] The information acquisition unit 139 acquires some or all of the application status information corresponding to the output instruction from the application management unit 111. For example, the information acquisition unit 139 acquires some or all of the application status information that is paired with the application identifier contained in the output instruction from the application management unit 111.

[0184] The transmission unit 14 transmits various types of information. These types of information include, for example, notification information, screen information, proposal content, negotiation information, necessary information, and agreement documents, which will be described later.

[0185] The petition notification unit 141, upon receiving petition information from the petition acquisition unit 121, acquires notification information containing the petitioner's information identified by the petitioner identifier in the petition information and the petition content information, and transmits the notification information to the respondent. The transmission of the information may be either push-type or pull-type transmission.

[0186] The screen transmission unit 142 transmits specific screen information to the petitioner terminal 2 or the respondent terminal 3 when the transmission conditions for transmitting specific screen information are met. For example, after confirmation information has been obtained, the screen transmission unit 142 transmits screen information to the respondent terminal 3 to allow the respondent to select one of two or more methods for processing the petition. For example, when the screen transmission unit 142 receives a selection of a selection object from the petitioner terminal 2 or the respondent terminal 3, it transmits screen information corresponding to that selection object to the petitioner terminal 2 or the respondent terminal 3.

[0187] The proposal content transmission unit 143 transmits the proposal content acquired by the proposal content acquisition unit 136 to the petitioner terminal 2 or the respondent terminal 3. The proposal content transmission unit 143 may transmit the same proposal content to both the petitioner terminal 2 and the respondent terminal 3. For example, in response to the receipt of a proposal content instruction, the proposal content transmission unit 143 transmits the proposal content corresponding to that instruction.

[0188] It is preferable for the proposal content transmission unit 143 to transmit the proposal content in such a way that the negotiation information received by the negotiation receiving unit 123 corresponds to the proposal content acquired by the proposal content acquisition unit 136. This is because the negotiation information that forms the basis of the proposal content can be grasped.

[0189] The negotiation information output unit 144 outputs two or more negotiation information received by the negotiation receiving unit 123. For example, the negotiation information output unit 144 outputs the two or more negotiation information received by the negotiation receiving unit 123 in the order in which it was received by the negotiation receiving unit 123, and in a manner that allows it to determine whether the identifier associated with the negotiation information is a petitioner identifier or a respondent identifier.

[0190] Here, "output" typically refers to transmission to petitioner terminal 2 or respondent terminal 3. However, "output" may also be a concept that includes display on a screen, projection using a projector, printing with a printer, sound output, transmission to an external device, storage on a recording medium, and delivery of processing results to other processing devices or programs other than petitioner terminal 2 or respondent terminal 3.

[0191] The necessary information transmission unit 145 transmits necessary information to the respondent, which is information necessary for the respondent to process the application, when the judgment unit 137 determines that the conditions of the agreement have been met. The necessary information usually includes payment destination information. Payment destination information may include, for example, bank account information, credit card information, and online payment information. Other necessary information may include, for example, information indicating the mailing address for the loaned items and information indicating the location where the service will be enjoyed.

[0192] The agreement document output unit 146 outputs the agreement document generated by the agreement document generation unit 138. The agreement document output unit 146 transmits the agreement document to the petitioner terminal 2 and the respondent terminal 3. It is preferable for the agreement document output unit 146 to transmit the agreement document corresponding to the instruction when an agreement document instruction is received.

[0193] The information transmission unit 147 transmits the information acquired by the processing unit 13 to the petitioner terminal 2 or the respondent terminal 3. The information transmission unit 147 transmits, for example, some or all of the petition status information acquired by the information acquisition unit 139 to the petitioner terminal 2 or the respondent terminal 3. The information transmission unit 147 transmits some or all of the petition status information to the petitioner terminal 2 or the respondent terminal 3 that has sent an output instruction. The information transmission unit 147 transmits, for example, the proposed content, the agreement document, or the payment schedule to the petitioner terminal 2 or the respondent terminal 3 that has sent a proposed content instruction, an agreement document instruction, or a payment schedule instruction.

[0194] The petitioner storage unit 21, which constitutes the petitioner terminal 2, stores various types of information. These types of information include, for example, the petitioner identifier.

[0195] The petitioner reception unit 22 receives various types of information and instructions. These include, for example, company notices, petition information, negotiation information, output instructions, proposal content instructions, agreement document instructions, or voice information from the petitioner. The petitioner reception unit 22 receives phone calls from the AI ​​agent or negotiation support device 1.

[0196] The means of inputting various information and instructions can be anything, such as a touch panel, keyboard, mouse, or menu screen.

[0197] The petitioner processing unit 23 performs various processes. These processes include, for example, converting received information and instructions into information and instructions for transmission. Other processes include, for example, converting received information into information for output.

[0198] The petitioner transmission unit 24 transmits various information and instructions to the negotiation support device 1 or other devices (not shown). These various information and instructions include, for example, company notices, petition information, negotiation information, or output instructions.

[0199] The applicant receiving unit 25 receives various types of information. These types of information include, for example, confirmation information, negotiation information, application status information, or various screen information.

[0200] The applicant output unit 26 outputs various types of information. These types of information include, for example, confirmation information, negotiation information, application status information, various screens, or voice information from the AI ​​agent.

[0201] Here, output typically includes concepts such as displaying on a screen, projecting using a projector, printing with a printer, sound output, transmission to an external device, storage on a recording medium, and handing over processing results to other processing devices or other programs.

[0202] The respondent storage unit 31, which constitutes the respondent terminal 3, stores various types of information. These types of information include, for example, the respondent identifier.

[0203] The respondent reception unit 32 receives various types of information and instructions. These types of information and instructions include, for example, confirmation information, negotiation information, output instructions, proposal content instructions, or agreement document instructions. The respondent reception unit 32 receives phone calls from the AI ​​agent or negotiation support device 1.

[0204] Any means of inputting information and instructions is acceptable, such as a microphone, touch panel, keyboard, mouse, or menu screen.

[0205] The respondent processing unit 33 performs various processes. These processes include, for example, converting received information and instructions into information and instructions for transmission. Other processes include, for example, converting received information into information for output.

[0206] The respondent's transmitting unit 34 transmits various information and instructions to the negotiation support device 1 or other devices (not shown). These various information and instructions include, for example, confirmation information, negotiation information, and output instructions.

[0207] The respondent's receiving unit 35 receives various types of information. These types of information include, for example, notification information, negotiation information, and debt status information.

[0208] The respondent output unit 36 ​​outputs various types of information. These types of information include, for example, notification information, negotiation information, debt status information, various screens, or voice information from the AI ​​agent.

[0209] Here, "output" is a concept that includes, for example, audio output from a speaker, display on a screen, projection using a projector, printing with a printer, transmission to an external device, storage on a recording medium, and transfer of processing results to other processing devices or other programs.

[0210] The storage unit 11, the application management unit 111, the condition management unit 112, the response information storage unit 113, the applicant storage unit 21, and the respondent storage unit 31 are preferably made of non-volatile recording media, but can also be made of volatile recording media.

[0211] The process by which information is stored in the storage unit 11, etc. is not relevant. For example, information may be stored in the storage unit 11, etc. via a recording medium, information transmitted via a communication line, etc. may be stored in the storage unit 11, etc., or information input via an input device may be stored in the storage unit 11, etc.

[0212] The receiving unit 12, the claim acquisition unit 121, the confirmation receiving unit 122, the negotiation receiving unit 123, the instruction receiving unit 124, the claimant receiving unit 25, and the respondent receiving unit 35 are usually implemented by wireless or wired communication means, but may also be implemented by means of receiving broadcasts.

[0213] The processing unit 13, petition storage unit 131, update unit 132, confirmation storage unit 133, negotiation support unit 134, negotiation storage unit 135, proposal content acquisition unit 136, judgment unit 137, agreement document generation unit 138, information acquisition unit 139, agent means 1341, device linkage means 13411, learning means 13412, response means 13413, petitioner processing unit 23, and respondent processing unit 33 can usually be implemented using a processor, memory, etc. The processing procedures for <components> etc. are usually implemented in software, and this software is recorded on a recording medium such as ROM. However, it may also be implemented in hardware (dedicated circuitry). The processor can be a CPU, MPU, GPU, etc., and the type is not specified.

[0214] The transmission unit 14, the petition notification unit 141, the screen transmission unit 142, the proposal content transmission unit 143, the negotiation information output unit 144, the necessary information transmission unit 145, the agreement document output unit 146, the information transmission unit 147, the petitioner transmission unit 24, and the respondent transmission unit 34 are usually implemented by wireless or wired communication means, but may also be implemented by broadcasting means.

[0215] The petitioner reception unit 22 and the respondent reception unit 32 can be implemented using device drivers for input means such as touch panels and keyboards, or control software for menu screens.

[0216] The petitioner's output unit 26 and the respondent's output unit 36 ​​may or may not be considered to include output devices such as displays and speakers. The respondent's output unit 36 ​​can be implemented using driver software for the output device, or driver software for the output device and the output device itself.

[0217] Next, the operation of the negotiation support device 1 will be explained using the flowcharts in Figures 4 to 6.

[0218] (Step S401) The receiving unit 12 determines whether or not it has received a company notification from the applicant's terminal 2. If it has received a company notification, it proceeds to step S402; otherwise, it proceeds to step S410.

[0219] (Step S402) The processing unit 13 obtains the received company notification.

[0220] (Step S403) The processing unit 13 obtains a unique claim identifier.

[0221] (Step S404) The processing unit 13 stores the company notification obtained in step S402 in the application management unit 111, associating it with the application identifier obtained in step S403. At this point, the update unit 132 may update the phase identification information to "company notification" in conjunction with the application identifier.

[0222] (Step S405) The decision unit 137 performs a decision process. An example of the decision process will be explained using the flowchart in Figure 7.

[0223] (Step S406) If the result of the decision in step S405 is "to perform dialogue processing", proceed to step S407; if it is "not to perform dialogue processing", proceed to step S408.

[0224] (Step S407) Negotiation support unit 134 performs dialogue processing. Return to step S401. An example of dialogue processing will be explained using the flowchart in Figure 8.

[0225] (Step S408) The processing unit 13 determines whether the notification conditions are met. If the notification conditions are met, the process proceeds to step S409; otherwise, it returns to step S401.

[0226] The notification conditions here include, for example, that the judgment result in step S405 is "no dialogue processing will be performed," and that the company's received notification is associated with the respondent's contact information. This contact information includes, for example, an email address, phone number, or social media ID.

[0227] (Step S409) The transmitting unit 14 sends the received company notice to the contact information of the respondent. Return to Step S401.

[0228] (Step S410) The petition acquisition unit 121 determines whether or not it has received the case registration from the petitioner's terminal 2. If the case registration has been received, the unit proceeds to step S411; otherwise, it proceeds to step S420.

[0229] (Step S411) The petition storage unit 131 obtains petition time information, which is the date and time of filing, from a clock (not shown). The petition storage unit 131 obtains petition information included in the case registration. The petition storage unit 131 is information having said petition information and said petition time information, and constitutes the petition information to be stored.

[0230] (Step S412) The claim storage unit 131 obtains a claim identifier. If a claim identifier is associated with the received case registration, the claim storage unit 131 obtains the claim identifier. If a claim identifier is not associated with the received case registration, the claim storage unit 131 obtains a unique claim identifier.

[0231] (Step S413) The application storage unit 131 stores the application information obtained in step S411 in the application management unit 111, paired with the application identifier obtained in step S412. At this point, the update unit 132 may update the phase identification information to "awaiting confirmation" paired with the application identifier.

[0232] (Step S414) The petition notification section 141 constitutes the notification information. An example of such notification information configuration process will be explained using the flowchart in Figure 10.

[0233] (Step S415) The decision unit 137 performs a decision process. An example of the decision process will be explained using the flowchart in Figure 7.

[0234] (Step S416) If the result of the decision in step S415 is "to perform dialogue processing", proceed to step S417; if it is "not to perform dialogue processing", proceed to step S418.

[0235] (Step S417) The negotiation support unit 134 performs dialogue processing. Return to Step S401. An example of dialogue processing will be explained using the flowchart in Figure 8.

[0236] (Step S418) The processing unit 13 determines whether the notification conditions are met. If the notification conditions are met, the process proceeds to step S419; otherwise, it returns to step S401.

[0237] The notification conditions here are, for example, that the judgment result in step S415 is "Do not perform dialogue processing", and that the respondent's contact information is associated with the received case registration.

[0238] (Step S419) The transmitting unit 14 sends the received company notice to the contact information of the respondent. Return to step S401.

[0239] (Step S420) The confirmation receiving unit 122 determines whether or not it has received confirmation information, etc. from the respondent terminal 3. If confirmation information, etc. has been received, the unit proceeds to step S421; otherwise, it proceeds to step S423. Confirmation information, etc. refers to confirmation information or non-existent information. The received confirmation information, etc. is associated with the claim identifier.

[0240] (Step S421) The confirmation storage unit 133 obtains the claim identifier associated with the confirmation information, etc.

[0241] (Step S422) The confirmation storage unit 133 obtains confirmation time information from a clock (not shown). The confirmation storage unit 133 stores the confirmation information, including the confirmation time information, in the application management unit 111, associating it with the application identifier obtained in step S421. The process returns to step S401. If the received information is confirmation information, the confirmation storage unit 133 may update the phase identification information associated with the application identifier to "Confirmed".

[0242] (Step S423) The negotiation receiving unit 123 determines whether or not it has received negotiation information from the petitioner's terminal 2 or the respondent's terminal 3. If negotiation information is received, the unit proceeds to step S424; otherwise, it proceeds to step S440. The received negotiation information is associated with the petition identifier.

[0243] (Step S424) The negotiation storage unit 135 obtains the claim identifier, etc. The claim identifier, etc., is, for example, the claim identifier and negotiation information.

[0244] (Step S425) The negotiation storage unit 135 stores the negotiation information in the application management unit 111, associating it with the application identifier obtained in step S424. The negotiation storage unit 135 may also obtain the inputter identifier and store the negotiation information including the inputter identifier.

[0245] (Step S426) The proposal content acquisition unit 136 determines whether the received negotiation information matches the proposal content creation conditions. If it matches the proposal content creation conditions, the process proceeds to step S427; otherwise, the process proceeds to step S429.

[0246] (Step S427) The proposal content acquisition unit 136 constructs the proposal content. An example of this proposal content construction process will be explained using the flowchart in Figure 11.

[0247] (Step S428) The proposal content acquisition unit 136 stores the proposal content, which was constructed in step S427, in the application management unit 111, associating it with negotiation information.

[0248] (Step S429) The determination unit 137 determines whether the received negotiation information matches the agreement conditions. If it matches the agreement conditions, the process proceeds to step S421; otherwise, it returns to step S426.

[0249] (Step S430) The agreement document generation unit 138 generates an agreement document. An example of such an agreement document generation process will be explained using the flowchart in Figure 12.

[0250] (Step S431) The agreement document generation unit 138 associates the agreement document acquired in step S430 with the petition identifier, and accumulates the agreement document in the petition management unit 111.

[0251] (Step S432) The processing unit 13 acquires the petitioner identifier. The required information transmission unit 145 acquires the required information paired with the petitioner identifier from the petition management unit 111. Note that the required information is, for example, payment destination information.

[0252] (Step S433) The processing unit 13 determines whether or not the required information has been acquired in step S423. If the required information has been acquired, the process proceeds to step S434, and if the required information has not been acquired, the process returns to step S401.

[0253] (Step S434) The required information transmission unit 145 transmits the required information to the respondent. The process returns to step S401. Note that the means for transmitting the required information here is not limited.

[0254] (Step S435) The determination unit 137 performs a determination process. An example of the determination process will be described using the flowchart of Fig. 7.

[0255] (Step S436) If the determination result in step S435 is "perform dialogue processing", the process proceeds to step S437, and if the determination result is "do not perform dialogue processing", the process proceeds to step S438.

[0256] (Step S437) The negotiation support unit 134 performs dialogue processing. The process returns to step S401. An example of the dialogue processing will be described using the flowchart of Fig. 8.

[0257] (Step S438) The processing unit 13 determines whether or not a notification condition is satisfied. If the notification condition is satisfied, the process proceeds to step S439, and if the notification condition is not satisfied, the process returns to step S401.

[0258] The notification conditions here are, for example, that the decision result in step S435 is "no dialogue processing," and that contact information of the opposing party (respondent or petitioner) that corresponds to the petition identifier associated with the received negotiation information exists in the petition management unit 111.

[0259] (Step S439) The transmitting unit 14 transmits the received negotiation information to the contact information of the other party. The process returns to Step S401.

[0260] (Step S440) The instruction receiving unit 124 determines whether or not it has received an output instruction from the petitioner terminal 2 or the respondent terminal 3. If an output instruction is received, the unit proceeds to step S441; otherwise, it proceeds to step S445. The output instruction is associated with the petition identifier.

[0261] (Step S441) The information acquisition unit 139 acquires the claim identifier corresponding to the output instruction received in step S440.

[0262] (Step S442) The information acquisition unit 139 acquires application status information from the application management unit 111 that corresponds to the application identifier acquired in step S441. Note that the application status information acquired here may be only a portion of the application status information from the application management unit 111.

[0263] (Step S443) The information acquisition unit 139 uses the application status information acquired in step S442 to construct the output information. An example of this output information configuration process will be explained using the flowchart in Figure 13.

[0264] (Step S444) The information transmission unit 147 transmits the output information to the terminal that sent the output instruction. Return to step S401.

[0265] (Step S445) The instruction receiving unit 124 determines whether or not it has received an instruction regarding the proposed content from the petitioner's terminal 2 or the respondent's terminal 3. If an instruction regarding the proposed content has been received, the unit proceeds to step S446; otherwise, it proceeds to step S448. The instruction regarding the proposed content corresponds to one piece of negotiation information.

[0266] (Step S446) The information acquisition unit 139 acquires from the claim management unit 111 a proposal that corresponds to one piece of negotiation information that corresponds to the proposal content instruction received in step S445.

[0267] (Step S447) The information transmission unit 147 transmits the proposal content obtained in step S446 to the terminal that sent the proposal content instruction. Return to step S401.

[0268] (Step S448) The instruction receiving unit 124 determines whether or not it has received an agreement document instruction from the petitioner terminal 2 or the respondent terminal 3. If an agreement document instruction has been received, the unit proceeds to step S449; otherwise, it proceeds to step S451. The agreement document instruction is associated with the petition identifier.

[0269] (Step S449) The information acquisition unit 139 obtains the claim identifier corresponding to the agreement document instruction received in step S448. The information acquisition unit 139 obtains the agreement document that is paired with the claim identifier from the claim management unit 111.

[0270] (Step S450) The information transmission unit 147 transmits the agreement document obtained in step S449 to the terminal that sent the agreement document instruction. The process returns to step S401.

[0271] (Step S451) The receiving unit 12 determines whether or not it has received payment information. If payment information has been received, it proceeds to step S452; otherwise, it proceeds to step S453. The payment information is associated with the claim identifier.

[0272] (Step S452) The processing unit 13 accumulates the payment information in association with the claim identifier. The process returns to step S401.

[0273] (Step S453) The processing unit 13 determines whether a transmission condition for transmitting specific screen information to the claimant terminal 2 or the respondent terminal 3 is satisfied. If the transmission condition is satisfied, the process proceeds to step S454, and if not, the process returns to step S401.

[0274] Note that the transmission condition is, for example, reception of confirmation information or specific negotiation information, or reception of a screen output instruction from the claimant terminal 2 or the respondent terminal 3. The transmission condition is associated with the screen information to be transmitted.

[0275] (Step S454) The information acquisition unit 139 acquires the screen information corresponding to the transmission condition from the storage unit 11. Note that the screen information corresponding to the transmission condition is, for example, screen information paired with the selected selected object.

[0276] (Step S455) The information transmission unit 147 transmits the screen information acquired in step S445 to the claimant terminal 2 or the respondent terminal 3. The process returns to step S401.

[0277] (Step S456) The negotiation support unit 134 substitutes 1 for a counter i.

[0278] (Step S457) The negotiation support unit 134 determines whether the i-th next timing information exists in the storage unit 11. If the i-th next timing information exists, the process proceeds to step S458, and if not, the process returns to step S401.

[0279] (Step S458) The negotiation support unit 134 determines whether the i-th next timing information is information indicating that it is currently the call timing. If it is the call timing, the process proceeds to step S459, and if it is not the call timing, the process proceeds to step S460.

[0280] (Step S459) The negotiation support unit 134 and agent means 1341 perform dialogue processing using the i-th next timing information, etc. An example of dialogue processing will be explained using the flowchart in Figure 8.

[0281] (Step S460) The negotiation support unit 134 increments counter i by 1.

[0282] In the flowcharts of Figures 4, 5, and 6, at some point, the learning means 13412 causes the AI ​​to learn the answer information stored in the answer information storage unit 113.

[0283] Furthermore, in the flowcharts of Figures 4, 5, and 6, processing is terminated by power-off or processing termination interrupts.

[0284] Next, we will explain examples of the decision-making processes in steps S405, S415, and S435 using the flowchart in Figure 7.

[0285] (Step S701) The determination unit 137 obtains a phase identifier from the storage unit 11 that identifies the current phase. Such a phase identifier is, for example, information associated with the application identifier.

[0286] (Step S702) The determination unit 137 obtains the received petitioner input or respondent input.

[0287] (Step S703) The determination unit 137 assigns 1 to counter i.

[0288] (Step S704) The determination unit 137 determines whether the i-th agent linkage condition exists in the condition management unit 112. If the i-th agent linkage condition exists, the process proceeds to step S705; otherwise, the process proceeds to step S709.

[0289] (Step S705) The decision unit 137 obtains the i-th agent cooperation condition from the condition management unit 112.

[0290] (Step S706) The determination unit 137 determines whether one or two types of information from the phase identifier obtained in step S701 or the input obtained in step S702 satisfy the i-th agent cooperation condition. If the i-th agent cooperation condition is satisfied, the unit proceeds to step S707; otherwise, the unit proceeds to step S708.

[0291] (Step S707) The decision unit 137 assigns "conversational processing (for example, "1")" to the variable "decision result". It returns to the higher-level processing.

[0292] (Step S708) The determination unit 137 increments counter i by 1. The process returns to step S704.

[0293] (Step S709) The decision unit 137 assigns "Do not perform interactive processing (for example, "0")" to the variable "Decision Result". It returns to the higher-level process.

[0294] Next, examples of the dialogue processing in steps S407, S417, S437, and S459 will be explained using the flowchart in Figure 8.

[0295] (Step S801) The agent means 1341 obtains from the prompt management unit 114 a prompt that corresponds to the agent cooperation condition or a prompt included in the next timing information.

[0296] (Step S802) Agent means 1341 obtains information to be given to the generating AI. Here, the information is, for example, one or more of the following: company notice, claim information, or negotiation information. Here, the information is, for example, information included in next timing information.

[0297] (Step S803) The agent means 1341 provides prompts, etc. to the generating AI. Prompts, etc. refer to the prompts obtained in step S801, or the prompts and information obtained in step S802.

[0298] (Step S804) Agent means 1341 obtains the telephone number of the other party. Agent means 1341 obtains, for example, a telephone number included in the company notice, claim information, negotiation information, or next timing information, or a telephone number of the respondent or claimant paired with a claim identifier paired with the company notice, claim information, negotiation information.

[0299] (Step S805) Agent means 1341 makes a call using the telephone number obtained in step S804.

[0300] (Step S806) Agent means 1341 determines whether the other party answered the phone. If the other party answered the phone, proceed to step S808; otherwise, proceed to step S807.

[0301] (Step S807) The agent means 1341 determines whether or not a timeout has occurred. If a timeout has occurred, the process proceeds to step S818; otherwise, it returns to step S806. A timeout occurs, for example, when a predetermined amount of time has elapsed since the call was made.

[0302] (Step S808) The agent means 1341 obtains a response from the generating AI. The response may be, for example, a string, but may also be voice information.

[0303] (Step S809) The agent means 1341 outputs the response from the generating AI as voice to the other party on the phone. For example, the agent means 1341 synthesizes the string response from the generating AI into voice, constructs voice information, and outputs the voice information over the phone.

[0304] (Step S810) The agent means 1341 determines whether or not it has received a response from the other party on the phone. If a response is received, it proceeds to step S811; otherwise, it returns to step S810.

[0305] (Step S811) The agent means 1341 stores the dialogue information in the claim management unit 111, associating it with the claim identifier. The dialogue information includes, for example, the response from the generating AI and the response from the other party on the phone. The responses from both parties included in the dialogue information may be voice information or speech-recognized strings.

[0306] (Step S812) The device linkage means 13411 performs device linkage processing. An example of device linkage processing will be explained using the flowchart in Figure 9.

[0307] (Step S813) Agent means 1341 determines whether the other party's response is an inquiry regarding the negotiation support device 1. If it is an inquiry, proceed to step S814; otherwise, proceed to step S816b.

[0308] The agent means 1341, for example, provides the generating AI with a query judgment prompt and the other party's response, and obtains information from the generating AI indicating whether or not the other party's response is an inquiry. The agent means 1341, for example, uses natural language processing to determine that the other party's response is an inquiry about the negotiation support device 1 if it contains predetermined cue phrases (for example, the name of the negotiation support device 1, "is it?").

[0309] (Step S814) The response means 13413 obtains an inquiry prompt from the prompt management unit 114.

[0310] (Step S815) The response means 13413 provides the generating AI with the inquiry prompt obtained in step S814 and the other party's response. The process returns to step S808.

[0311] (Step S816) Agent means 1341 decides whether to terminate the dialogue or not. If it decides to terminate the dialogue, it returns to the higher-level process; otherwise, it proceeds to step S817.

[0312] Agent 1341 determines to terminate the conversation, for example, when the variable "cooperation result" is assigned "conversation ended" (e.g., "-1"), when it receives a call disconnection signal, or when it detects that the response contains information indicating the end of the conversation. Agent 1341 also determines to terminate the conversation, for example, when it receives information indicating "conversation ended" from the generating AI that provided the inquiry judgment prompt. If it determines to terminate the conversation, agent 1341 then performs the process of terminating the conversation with the user.

[0313] (Step S817) Agent means 1341 gives the other party's response to the generating AI. Return to step S808.

[0314] (Step S818) The agent means 1341 acquires next timing information. Next timing information is information that identifies the timing of the next call. For example, the next timing information is a predetermined time from the present. For example, the next timing information is a predetermined time from the present and between 10:00 and 18:00 (a time when people are usually active).

[0315] (Step S819) The agent means 1341 stores the next timing information, etc., associated with the claim identifier. It returns to the higher-level processing. The next timing information, etc., includes, for example, the next timing information, the other party's telephone number, and the information given to the generating AI in S803. The next timing information, etc., includes, for example, the prompt or prompt identifier given to the generating AI immediately before.

[0316] In addition, in the flowchart of Figure 8, it is preferable that the next timing information is deleted when the stored next timing information is used to make a call and the conversation begins.

[0317] Next, an example of the device linkage process in step S812 will be explained using the flowchart in Figure 9.

[0318] (Step S901) The device linkage means 13411 obtains a phase identifier that identifies the current phase.

[0319] (Step S902) The device linking means 13411 assigns 1 to counter i.

[0320] (Step S903) The device linkage means 13411 determines whether the i-th device linkage condition exists in the condition management unit 112. If the i-th device linkage condition exists, the process proceeds to step S904; otherwise, it returns to the higher-level process.

[0321] (Step S904) The device linkage means 13411 obtains the i-th device linkage condition from the condition management unit 112.

[0322] (Step S905) The device linkage means 13411 obtains one or more pieces of information to determine the i-th device linkage condition. Such information may be, for example, a phase identifier, a claimant input, or a respondent input.

[0323] (Step S906) The device linking means 13411 determines whether one or more pieces of information satisfy the i-th device linking condition. If the i-th device linking condition is satisfied, proceed to step S907; otherwise, proceed to step S911.

[0324] (Step S907) The device linkage means 13411 obtains a processing identifier that corresponds to the i-th device linkage condition from the condition management unit 112.

[0325] (Step S908) The device linkage means 13411 executes the process identified by the process identifier.

[0326] (Step S909) The device linkage means 13411 determines whether a dialogue termination flag corresponding to the i-th device linkage condition exists in the condition management unit 112. If the dialogue termination flag exists, the process proceeds to step S910; otherwise, the process proceeds to step S911.

[0327] (Step S910) The device linking means 13411 assigns "end of dialogue (for example, "-1")" to the variable "linking result".

[0328] (Step S911) The device linkage means 13411 increments counter i by 1. Return to step S903.

[0329] Next, an example of the notification information configuration process in step S414 will be explained using the flowchart in Figure 10.

[0330] (Step S1001) The petition notification unit 141 obtains a notification information template from the storage unit 11.

[0331] (Step S1002) The petition notification unit 141 obtains a petition identifier. The petition notification unit 141 obtains information about the petitioner (for example, the name of the petitioner) that is paired with the petition identifier.

[0332] (Step S1003) The petition notification section 141 substitutes the petitioner information obtained in step S1002 into the variable that is used to substitute the petitioner's information in the notification information template.

[0333] (Step S1004) The petition notification unit 141 obtains information about the respondent (for example, the name of the respondent) that is paired with the petition identifier.

[0334] (Step S1005) The petition notification section 141 substitutes the respondent's information into the variable in the notification information template that is used to substitute the respondent's information.

[0335] (Step S1006) The petition notification unit 141 obtains link information for accessing the petition content information that is paired with the petition identifier. The link information is usually a URL, but any information that allows access to the petition content information is acceptable.

[0336] (Step S1007) The petition notification unit 141 assigns the link information to the variable that is assigned the link information in the notification information template. It then returns to the higher-level process.

[0337] Next, an example of the proposed content configuration process in step S427 will be explained using the flowchart in Figure 11.

[0338] (Step S1101) The proposal content acquisition unit 136 acquires a proposal content template from the storage unit 11.

[0339] The proposal content acquisition unit 136, for example, acquires a claim processing identifier included in the negotiation information and acquires a proposal content template that corresponds to the said claim processing identifier from the storage unit 11.

[0340] (Step S1102) The proposal content acquisition unit 136 acquires the claim identifier corresponding to the received negotiation information.

[0341] (Step S1103) The proposal content acquisition unit 136 assigns 1 to counter i.

[0342] (Step S1104) The proposal content acquisition unit 136 determines whether the i-th variable exists in the proposal content template acquired in step S1101. If the i-th variable exists, the process proceeds to step S1105; otherwise, it returns to the higher-level process.

[0343] (Step S1105) The proposal content acquisition unit 136 acquires information corresponding to the i-th variable from the application status information paired with the application identifier.

[0344] (Step S1106) The proposal content acquisition unit 136 assigns the information acquired in step S1105 to the i-th variable in the proposal content template.

[0345] (Step S1107) The proposal content acquisition unit 136 increments counter i by 1. Return to step S1104.

[0346] Next, an example of the agreement document generation process in step S430 will be explained using the flowchart in Figure 12.

[0347] (Step S1201) The agreement document generation unit 138 obtains an agreement document template from the storage unit 11.

[0348] (Step S1202) The agreement document generation unit 138 obtains the claim identifier corresponding to the received negotiation information.

[0349] (Step S1203) The agreement document generation unit 138 assigns 1 to counter i.

[0350] (Step S1204) The agreement document generation unit 138 determines whether the i-th variable exists in the agreement document template obtained in step S1201. If the i-th variable exists, the unit proceeds to step S1205; otherwise, it returns to the higher-level process.

[0351] (Step S1205) The agreement document generation unit 138 obtains information corresponding to the i-th variable from the application status information paired with the application identifier.

[0352] (Step S1206) The agreement document generation unit 138 substitutes the information obtained in step S1205 into the agreement document template at the location of the i-th variable.

[0353] (Step S1207) The agreement document generation unit 138 increments counter i by 1. Return to step S1204.

[0354] Next, an example of the output information configuration process in step S443 will be explained using the flowchart in Figure 13.

[0355] (Step S1301) The information acquisition unit 139 acquires screen information from the storage unit 11 that will be used as the basis for constructing the output information.

[0356] (Step S1302) The information acquisition unit 139 assigns 1 to counter i.

[0357] (Step S1303) The information acquisition unit 139 determines whether or not the i-th variable exists in the screen information. If the i-th variable exists, the unit proceeds to step S1304; otherwise, the unit proceeds to step S1307.

[0358] (Step S1304) The information acquisition unit 139 acquires the petition identifier. The information acquisition unit 139 acquires the information corresponding to the i-th variable in the screen information from the petition status information that is paired with the petition identifier. The information corresponding to the i-th variable is, for example, the name of the petitioner, the name of the respondent, the payment amount, the payment start date, and the payment due date.

[0359] (Step S1305) The information acquisition unit 139 assigns the information acquired in step S1304 to the i-th variable in the screen information.

[0360] (Step S1306) The information acquisition unit 139 increments counter i by 1. Return to step S1303.

[0361] (Step S1307) The information acquisition unit 139 acquires application-related information using the application information in the application status information that is paired with the application identifier acquired in step S1304. The application-related information includes, for example, information indicating that an application has been filed (e.g., "I have filed an application."), application time information (e.g., year, month, day and time), and the name of the applicant.

[0362] (Step S1308) The information acquisition unit 139 determines whether or not confirmation information exists in the application status information that is paired with the application identifier acquired in step S1304. If confirmation information exists, the process proceeds to step S1309; otherwise, it returns to the higher-level process.

[0363] (Step S1309) The information acquisition unit 139 acquires confirmation-related information using confirmation information of the application status information that is paired with the application identifier. The confirmation-related information includes, for example, information indicating that the application has been confirmed (e.g., "The case has been confirmed."), confirmation time information (e.g., year, month, and time), and the name of the respondent.

[0364] (Step S1310) The information acquisition unit 139 assigns 1 to counter i.

[0365] (Step S1311) The information acquisition unit 139 determines whether or not the i-th negotiation information exists in the application status information that is paired with the application identifier acquired in step S1304. If the i-th negotiation information exists, the unit proceeds to step S1312; otherwise, the unit proceeds to step S1319.

[0366] (Step S1312) The information acquisition unit 139 acquires the petitioner identifier or respondent identifier that is paired with the i-th negotiation information from the petition status information that is paired with the said petition identifier. The petitioner identifier or respondent identifier will be referred to as identifiers as appropriate.

[0367] (Step S1313) The information acquisition unit 139 determines whether the proposed content is associated with the i-th negotiation information. If the proposed content is associated, the process proceeds to step S1314; otherwise, the process proceeds to step S1315.

[0368] (Step S1314) The information acquisition unit 139 uses the i-th negotiation information and the proposed content to construct the negotiation relationship information to be output. Proceed to step S1316. Note that the negotiation relationship information here includes, for example, a sentence explaining the negotiation information, negotiation time information, and link information to the proposed content.

[0369] (Step S1315) The information acquisition unit 139 uses the i-th negotiation information to construct the negotiation relationship information to be output. The negotiation relationship information here includes, for example, a sentence explaining the negotiation information and negotiation time information.

[0370] (Step S1316) The information acquisition unit 139 places the i-th negotiation relationship information below the information already placed, where the identifier is located (for example, the right or left side of the screen).

[0371] (Step S1317) The information acquisition unit 139 increments counter i by 1. Return to step S1311.

[0372] (Step S1318) The information acquisition unit 139 determines whether an agreement has been reached. If an agreement has been reached, it proceeds to step S1319; otherwise, it returns to the higher-level process.

[0373] (Step S1319) The information acquisition unit 139 acquires the necessary information that corresponds to the claim identifier. The necessary information is usually the recipient information.

[0374] (Step S1320) The information acquisition unit 139 places the necessary information on the screen. It then returns to the higher-level processing.

[0375] Next, we will explain an example of the operation of the petitioner's terminal 2 using the flowchart in Figure 14.

[0376] (Step S1401) The petitioner reception department 22 determines whether or not it has received the company's notification. If the company's notification has been received, it proceeds to step S1402; if it has not been received, it proceeds to step S1403.

[0377] (Step S1402) The petitioner transmission unit 24 transmits the company notice received in step S1401 to the negotiation support device 1. Return to step S1401.

[0378] (Step S1403) The petitioner reception department 22 determines whether or not the case registration has been accepted. If the case registration has been accepted, the process proceeds to step S1404; otherwise, the process proceeds to step S1405.

[0379] (Step S1404) The petitioner transmission unit 24 transmits the case registration received in step S1403 to the negotiation support device 1. Return to step S1401.

[0380] (Step S1405) The petitioner reception unit 22 determines whether or not it has received the instructions. If the instructions have been received, it proceeds to step S1406; otherwise, it proceeds to step S1409. The instructions may include, for example, output instructions, screen output instructions, or other instructions.

[0381] (Step S1406) The petitioner processing unit 23 obtains the petition identifier. The petitioner transmission unit 24 transmits the instructions received in step S1405 to the negotiation support device 1, associating them with the petition identifier.

[0382] (Step S1407) The petitioner receiving unit 25 determines whether or not it has received information corresponding to the instructions. If it has received information, it proceeds to step S1408; otherwise, it returns to step S1407.

[0383] (Step S1408) The petitioner processing unit 23 configures the information to be output from the received information. The petitioner output unit 26 outputs the said information. The process returns to step S1401.

[0384] (Step S1409) The petitioner reception department 22 determines whether or not it has received the negotiation information. If it has received the negotiation information, it proceeds to step S1410; if it has not received it, it proceeds to step S1411.

[0385] (Step S1410) The petitioner processing unit 23 obtains the petition identifier. The petitioner transmission unit 24 transmits the negotiation information received in step S1409 to the negotiation support device 1, associating it with the petition identifier. The process returns to step S1401.

[0386] (Step S1411) The petitioner's receiving unit 25 determines whether or not a call has been received. If a call has been received, the unit proceeds to step S1412; otherwise, it proceeds to step S1413.

[0387] (Step S1412) The petitioner processing unit 23 activates the telephone function. From this point on, the petitioner interacts with the AI ​​agent by telephone. Return to Step S1401.

[0388] (Step S1413) The petitioner reception department 22 determines whether or not it has received the instruction to disconnect the phone. If it has received the instruction to disconnect the phone, it proceeds to step S1414; otherwise, it proceeds to step S1415.

[0389] (Step S1414) The petitioner processing unit 23 performs the telephone disconnection process. Return to Step S1401.

[0390] (Step S1415) The petitioner receiving unit 25 determines whether or not it has received information from the negotiation support device 1. If information has been received, it proceeds to step S1416; otherwise, it returns to step S1401.

[0391] (Step S1416) The petitioner processing unit 23 configures the information to be output from the received information. The petitioner output unit 26 outputs the said information. The process returns to step S1401.

[0392] In the flowchart in Figure 14, processing is terminated by power-off or processing termination interrupts.

[0393] Next, we will explain an example of the operation of the respondent's terminal 3 using the flowchart in Figure 15.

[0394] (Step S1501) The respondent receiving unit 35 determines whether or not it has received a company notification from the negotiation support device 1. If it has received a company notification, it proceeds to step S1502; otherwise, it proceeds to step S1503. The received company notification is associated with a claim identifier.

[0395] (Step S1502) The respondent output unit 36 ​​outputs the company's notification received in step S1501.

[0396] (Step S1503) The respondent receiving unit 35 determines whether or not it has received notification information from the negotiation support device 1. If it has received notification information, it proceeds to step S1504; otherwise, it proceeds to step S1507. The received notification information is associated with a claim identifier.

[0397] (Step S1504) The respondent output unit 36 ​​outputs the notification information received in step S1503.

[0398] (Step S1505) The respondent reception unit 32 determines whether or not it has received the confirmation information, etc. If it has received the confirmation information, etc., it proceeds to step S1506; otherwise, it returns to step S1505.

[0399] (Step S1506) The respondent processing unit 33 obtains the claim identifier associated with the notification information. The respondent transmission unit 34 transmits the confirmation information, etc., associated with the claim identifier to the negotiation support device 1. The process returns to step S1501.

[0400] (Step S1507) The respondent reception department 32 determines whether or not it has received the instructions. If the instructions have been received, it proceeds to step S1508; if they have not been received, it proceeds to step S1511.

[0401] (Step S1508) The respondent processing unit 33 obtains a claim identifier corresponding to the screen on which instructions are entered. The respondent transmission unit 34 transmits the instructions to the negotiation support device 1, associating them with the claim identifier.

[0402] (Step S1509) The respondent's receiving unit 35 determines whether or not it has received information corresponding to the instruction. If it has received information, it proceeds to step S1510; otherwise, it returns to step S1509.

[0403] (Step S1510) The respondent processing unit 33 constructs the output information from the information received in step S1509. The respondent output unit 36 ​​outputs the said information. The process returns to step S1501.

[0404] (Step S1511) The respondent reception department 32 determines whether or not it has received the negotiation information. If the negotiation information has been received, it proceeds to step S1512; otherwise, it proceeds to step S1513.

[0405] (Step S1512) The respondent processing unit 33 obtains a claim identifier corresponding to the screen for inputting negotiation information. The respondent transmission unit 34 transmits the negotiation information received in step S1511 to the negotiation support device 1, associating it with the said claim identifier. The process returns to step S1501.

[0406] (Step S1513) The respondent reception department 32 determines whether or not the call was received. If the call was received, proceed to step S1514; if the call was not received, proceed to step S1515.

[0407] (Step S1514) The respondent processing unit 33 activates the telephone function. The process returns to Step S1501. From this point onward, the respondent will interact with the AI ​​agent.

[0408] (Step S1515) The respondent reception unit 32 determines whether or not it has received the instruction to disconnect the phone. If it has received the instruction to disconnect the phone, it proceeds to step S1516; if it has not received the instruction to disconnect the phone, it proceeds to step S1517.

[0409] (Step S1516) The respondent processing unit 33 performs the telephone disconnection process. Return to Step S1501.

[0410] (Step S1517) The respondent receiving unit 35 determines whether or not it has received information from the negotiation support device 1. If information has been received, it proceeds to step S1518; otherwise, it returns to step S1516.

[0411] (Step S1518) The respondent processing unit 33 configures the information to be output from the received information. The respondent output unit 36 ​​outputs the said information. Return to step S1501.

[0412] In the flowchart shown in Figure 15, processing is terminated by power-off or processing termination interrupts.

[0413] The following describes a specific example of the operation of information system A in this embodiment. Here, the application handled by negotiation support device 1 is assumed to be a monetary application. Figure 16 shows the flow of phases in negotiation support device 1 from the company's notification of the monetary application to the occurrence of the application, and from the agreement between the applicant (creditor) and the respondent (debtor) regarding the processing of the application to the completion of the application processing. In Figure 16, the phases proceed in the order of "Company Notification," "Application Filed," "Waiting for Confirmation," "Confirmed," "Negotiating," "Application Processing," and "Application Processing Completed." Here, it is assumed that during "Negotiating," the process may transition to, for example, "Online Mediation." Furthermore, from "Online Mediation," the phase progresses to "Application Processing" or "Negotiation Failed." The numbers above the strings indicating each phase in Figure 16 are phase identifiers that identify the phase. In other words, the phase identifier for phase "Notification from the Company" is "0", the phase identifier for phase "Application Filed" is "1", "Awaiting Confirmation" is "2", "Confirmed" is "3", "Negotiating" is "4", "Application Processing" is "5", "Application Processing Completed" is "6", "Online Mediation in Progress" is "7", and "Negotiations Failed" is "8". Note that "Application Processing" refers to, for example, the stage of payments in a multi-installment payment plan.

[0414] The storage unit 11 of the negotiation support device 1 stores the petitioner management table shown in Figure 17. The petitioner management table is a table that manages information about petitioners. The petitioner management table is a table that manages petitioner information registered by the user in order to receive the services of the negotiation support device 1. The petitioner management table manages one or more records that have "ID", "petitioner identifier", "petitioner name", "email address", and "payment destination information". Note that "petitioner name", "email address", and "payment destination information" constitute petitioner information. "Payment destination information" is, for example, bank account information.

[0415] The application management unit 111 of the negotiation support device 1 stores an application management table having the structure shown in Figure 18. The application management table is a table that manages application status information. The application management table is a table that manages one or more records having "Application Identifier," "Application Information," "Confirmation Information," "Application Processing Identifier," "Phase Identifier," "Negotiation Information," "Time of Negotiation," "Proposed Content," and "Agreement Document." "Application Information" has "Application Content Information," "Applicant Identifier," "Respondent Information," and "Time of Application." "Confirmation Information" has "Time of Confirmation." "Negotiation Information" here has "Inputter" and "Negotiation Content." "Inputter" has either "1," indicating that the person who input the negotiation information is the applicant, or "2," indicating that the person who input the negotiation information is the respondent.

[0416] Regarding the information that constitutes the "Application Processing Identifier," here we will assign the following to each category: "Pay now by credit card" is "1," "Pay in full by the end of next month" is "2," "Pay in installments at the end of each month" is "3," "Proposal to revise the amount" is "4," "I have no recollection of the claim" is "5," "Already paid directly" is "6," and "Request online mediation" is "7." The "Application Processing Identifier" may also have a payment method identifier. For example, the payment method identifier could be "A" for credit card payment or "B" for bank transfer.

[0417] The storage unit 11 of the negotiation support device 1 stores a template of the proposed content for each application processing identifier (Figure 19). The storage unit 11 of the negotiation support device 1 also stores a template of the agreement document for each application processing identifier (Figure 20). In the template of the proposed content and the template of the agreement document, strings enclosed in "<" and ">" are variables. The information in the variables is replaced, and the proposed content is constructed. The variable <Respondent Name> is assigned the respondent's name, and the variable <Petitioner Name> is assigned the petitioner's name.

[0418] In the above situation, the following two specific examples will be explained. Specific example 1 is the case where the prompt given to the generating AI contains both device linkage conditions and a processing identifier. Specific example 2 is the case where the device linkage conditions and the processing identifier are managed separately from the prompt.

[0419] (Specific Example 1) The prompt management unit 114 of the negotiation support device 1 stores prompts associated with agent cooperation conditions (see FIG. 21 and FIG. 22). The prompts in FIG. 21 and FIG. 22 are prompts for causing generative AI to function as an AI agent. The agent cooperation condition in FIG. 21 is "that a company notification has been received (also referred to as 'phase identifier = company notification')", and the agent cooperation condition in FIG. 22 is "that petition information has been received (also referred to as 'phase identifier = petition completed')". It should be noted that when a company notification is received, the phase identifier "company notification" is set. Also, when petition information is received, the phase identifier "petition completed" is set.

[0420] In the prompt of FIG. 21, 2101 and 2103 are device cooperation conditions, and 2102 and 2104 are process identifiers.

[0421] In the above situation, it is assumed that a petitioner who has sold a product to the respondent and has not received payment from the respondent inputs a company notification "<Invoice> Invoice 1.pdf <Respondent Information> <Name> Company X <Full Name> Yamada Yuko <Respondent. Contact Information> 090-1234-5566 ···< / Respondent Information>···" into the petitioner terminal 2. It is assumed that the billed amount in "Invoice 1.pdf" is "100,000 JPY".

[0422] Next, the petitioner terminal 2 accepts the company notification, and transmits the company notification to the negotiation support device 1.

[0423] Next, the receiving unit 12 of the negotiation support device 1 receives the company notification from the petitioner terminal 2. It is assumed that the company notification is associated with the petitioner identifier "C01".

[0424] Next, the processing unit 13 acquires the received company notification. Next, the processing unit 13 generates a unique claim identifier "1". Next, the processing unit 13 stores the company notification in the claim management unit 111 in association with the claim identifier. Here, the update unit 132 stores the phase identification information "company notification '0'" in pairs with the claim identifier.

[0425] Next, the determination unit 137 performs the determination processing described with reference to the flowchart in Figure 7. That is, the determination unit 137 acquires the phase identifier "company notification". The determination unit 137 acquires the received company notification which is the claimant input, "<Invoice> Invoice 1.pdf <Respondent Information> <Name> Company X <Full Name> Yamada Yoko <Respondent. Contact Information> 090-1234-5566 ··· < / Respondent Information> ···". Next, the determination unit 137 determines that the phase identifier "company notification" satisfies the agent cooperation condition in Figure 21.

[0426] Next, the negotiation support unit 134 performs the interaction processing described with reference to the flowchart in Figure 8. First, the agent unit 1341 acquires from the prompt management unit 114 the prompt (Figure 21) paired with the agent cooperation condition (phase identifier = company notification). In addition, the agent unit 1341 acquires the company notification which is the information to be provided to the generative AI. Next, the agent unit 1341 substitutes "<Invoice> Invoice 1.pdf <Respondent Information> <Name> Company X <Full Name> Yamada Yoko <Respondent. Contact Information> 090-1234-5566 ··· < / Respondent Information> ···" into the variable <company notification> of the acquired prompt, and substitutes "090-1234-5566" into the variable <Respondent. Contact Information>. Further, the agent unit 1341 acquires the payment destination information "Bank Account A" paired with the claimant identifier "C01" from the claimant management table (Figure 17), and substitutes it into the variable <Claimant Identifier. Payment Destination Information> of the prompt. Next, the agent unit 1341 provides the prompt with information substituted into each variable to the generative AI.

[0427] The generating AI then dials the respondent's phone number, "090-1234-5566," and the AI ​​agent based on the generating AI converses with Yamada Y. of Company X, the respondent, by phone, following the prompts in Figure 21. The processing of the generating AI can also be considered as the processing of agent means 1341.

[0428] The respondent then stated over the phone that he could pay in installments and that the number of installments should be 20. The AI ​​agent then uses the billing amount of 100,000 yen and the number of installments of 20 in Invoice1.pdf to obtain the installment amount of 5,000 yen.

[0429] Next, since the respondent's statement (input) matches the device linkage conditions of 2103, the AI ​​agent executes the process "Send_message($petitioner.contact information,$acquired information)" identified by the processing identifier of 2104. Here, the variable "$petitioner.contact information" is assigned the email address of the petitioner "Tanaka A-ko" "ta@c.com", and the variable "$acquired information" is assigned the information that "payment can be made in installments", the number of installments "20", and the installment amount "5000 yen". Then, the AI ​​agent executes "Send_message(ta@c.com,“Payment can be made in installments <number of installments>20 <installment amount>5000 yen”)". As a result, the information “Payment can be made in installments <number of installments>20 <installment amount>5000 yen” is sent to Tanaka A-ko via email.

[0430] Let's say Tanaka A-ko, after viewing information regarding installment payments from the respondent, logs into negotiation support device 1 and enters her desired installment terms. In other words, for example, let's say Tanaka A-ko's petitioner terminal 2 accesses negotiation support device 1, and on the screen shown in Figure 23 displayed on petitioner terminal 2, she clicks the "Propose a change in installment amount" button 2301. Then, petitioner terminal 2 receives this instruction and sends a screen output instruction for the "Propose a change in installment amount" button 2301 to negotiation support device 1.

[0431] Next, the instruction receiving unit 124 of the negotiation support device 1 receives the screen output instruction. Next, the information acquisition unit 139 acquires the screen information corresponding to the screen output instruction. Next, the information transmission unit 147 transmits the screen information to Tanaka A's petitioner terminal 2.

[0432] Next, Tanaka A's petitioner terminal 2 receives the screen information and outputs the installment amount change screen shown in Figure 24. Then, Tanaka A enters the proposed new installment amount "20,000 yen" on the screen in Figure 24 (Figure 24, 2401).

[0433] Next, the petitioner's terminal 2 receives the proposed new installment amount of "20,000 yen" and transmits the negotiation information, including the said installment amount of "20,000 yen," to the negotiation support device 1, paired with the petition identifier "1."

[0434] Next, the negotiation receiving unit 123 of the negotiation support device 1 receives negotiation information including the installment amount "20,000 yen" paired with the application identifier "1".

[0435] Next, the proposal content acquisition unit 136 determines that the received negotiation information meets the proposal content creation condition "includes installment payments".

[0436] Next, the proposal content acquisition unit 136 performs the proposal content configuration process described using the flowchart in Figure 11. Specifically, the proposal content acquisition unit 136 acquires a proposal content template that corresponds to the proposal content creation conditions from the storage unit 11, and substitutes the installment amount "20,000 yen," the petitioner's name, the respondent's name, and the petition amount that correspond to the petition identifier "1" into the proposal content template to configure the proposal content. Next, the transmission unit 14 transmits the proposal content to Tanaka A's petitioner terminal 2.

[0437] Next, the negotiation storage unit 135 obtains the claim identifier "1" which is paired with the negotiation information. The negotiation storage unit 135 also obtains the time information "2024 / 8 / 11 16:34" from a clock (not shown). The negotiation storage unit 135 also obtains the claim amount "100,000 yen" and the installment amount "20,000 yen" which are paired with the claim identifier "1", and obtains the payment start date, payment end date, and payment amount for each month (obtains the payment schedule), and stores them paired with the claim identifier "1".

[0438] Next, the proposal content acquisition unit 136 acquires the previously acquired proposal content. Then, the proposal content acquisition unit 136 associates the proposal content with the negotiation information (file "P2") and stores it in the application management table (Figure 18).

[0439] Next, the processing unit 13 determines that the transmission condition for sending specific screen information, "the proposed content has been configured," is met. Next, the information acquisition unit 139 configures screen information including the negotiation information and link information to the proposed content. Next, the information transmission unit 147 transmits the screen information to Yamada Y's respondent terminal 3. The information transmission unit 147 may also transmit a link (URL) to the screen information to Yamada Y's respondent terminal 3.

[0440] Next, the respondent terminal 3 receives the screen information and outputs the screen shown in Figure 25. Then, Yamada Y. instructs the user to press the "Accept the revised installment amount" button 2501 in Figure 25. The respondent terminal 3 then accepts the instruction to press the button 2501. Next, the respondent terminal 3 obtains the information "<Negotiation Result> Agreement" associated with the button 2501, constructs negotiation information "<Petition Identifier>1 <Negotiation Result> Agreement <Installment Amount>20000 <Respondent Identifier>Company X" containing this information, and transmits it to the negotiation support device 1.

[0441] Next, the negotiation receiving unit 123 of the negotiation support device 1 receives the negotiation information. The negotiation storage unit 135 obtains the claim identifier "1". The negotiation storage unit 135 obtains the current time, the time of agreement "2024 / 8 / 11 16:39", from a clock (not shown). Next, the negotiation storage unit 135 stores the negotiation information "agreement" and the time of agreement in association with the claim identifier "1".

[0442] Next, the judgment unit 137 determines that the received negotiation information matches the agreement condition "The negotiation information includes '<negotiation result> agreement'." Then, the agreement document generation unit 138 generates the agreement document "A1" using the agreement document generation process explained using the flowchart in Figure 12. The agreement document generation unit 138 then stores the agreement document "A1" in the application management table (Figure 18) paired with the application identifier "1."

[0443] Next, the information acquisition unit 139, in response to the respondent's (Yoko Yamada) button 2501 (Figure 25), acquires screen information corresponding to the negotiation information "<Petition Identifier>1 <Negotiation Result>Agreement <Installment Amount>20000 <Respondent Identifier>Company X", which is to be sent to the petitioner. Next, the information transmission unit 147 transmits this screen information to the petitioner's terminal 2.

[0444] Furthermore, the information acquisition unit 139 acquires screen information that allows the respondent to determine the petition processing identifier, in response to the respondent's button 2501 (Figure 25). Next, the information transmission unit 147 transmits this screen information to the respondent's terminal 3.

[0445] Next, the petitioner's terminal 2 receives the screen information transmitted by the information transmission unit 147 and outputs the screen shown in Figure 26. The screen in Figure 26 is an example of a screen showing the process of filing the petition, confirming the case, and negotiations leading to an agreement.

[0446] Furthermore, the respondent terminal 3 receives screen information for determining the petition processing identifier and outputs a screen for determining the petition processing identifier. An example of such a screen is shown in Figure 27.

[0447] Next, it is stated that Yamada Y. instructed the respondent terminal 3 to click the "Pay by bank transfer" button 2701 on the screen shown in Figure 27.

[0448] Next, the respondent terminal 3 receives the instruction and obtains the application processing identifier "B". It should be noted that application processing identifier "A" represents "payment by credit card" and "B" represents "payment by bank transfer". Next, the respondent terminal 3 transmits the application processing identifier "B" to the negotiation support device 1, paired with application identifier "1".

[0449] Next, the negotiation support device 1 receives the application processing identifier "B," etc., and stores the application processing identifier "B" in the application management table (Figure 18) in conjunction with the application identifier "1." As a result of this process, the application processing identifier "3,B" is stored in the record "ID=1" in the application management table.

[0450] Furthermore, the update unit 132 of the negotiation support device 1 updates the phase identifier, which is paired with the claim identifier "1", to "5".

[0451] Next, the information acquisition unit 139 acquires the petitioner identifier "C01," which is paired with the petitioner identifier "1," from the petition management table (Figure 18). Next, the information acquisition unit 139 acquires the deposit destination information "Bank Account A," which is paired with the petitioner identifier "C01." Next, the information acquisition unit 139 constructs screen information including the said deposit destination information "Bank Account A." Next, the information transmission unit 147 transmits the said screen information to the respondent terminal 3.

[0452] Next, the respondent's terminal 3 receives the screen information and outputs the screen shown in Figure 28. In Figure 28, 2801 is the information for "Bank Account A".

[0453] (Specific example 2) The condition management unit 112 of the negotiation support device 1 stores the agent linkage condition management table shown in Figure 29. The agent linkage condition management table is a table that manages agent linkage conditions and prompts that correspond to those agent linkage conditions. The agent linkage condition management table is a table that manages one or more records that have an "agent linkage condition" and a "prompt".

[0454] Furthermore, the condition management unit 112 stores the device linkage condition management table shown in Figure 30. The device linkage condition management table is a table that manages device linkage conditions and the processing identifiers corresponding to those device linkage conditions. The device linkage condition management table is a table that manages one or more records that have "device linkage conditions" and "processing identifiers". In this case, "device linkage conditions" include "phase identifier conditions" and "input conditions". "Phase identifier conditions" are conditions related to phases. "Input conditions" are conditions related to applicant input or respondent input.

[0455] In the circumstances described above, Tanaka A, who has registered her petitioner information in negotiation support device 1, logged into negotiation support device 1 using petitioner terminal 2.

[0456] Then, through the actions of Tanaka A, the petitioner's terminal 2 received screen information for filing a claim from the negotiation support device 1 and output the case registration screen shown in Figure 31.

[0457] Next, Tanaka A entered information into each field on the screen shown in Figure 31, such as "Petition Name," "Petition Amount," and "Payment Due Date," and then clicked the "Submit Petition" button 2101.

[0458] Next, the petitioner reception unit 22 of the petitioner terminal 2 accepts the case registration in accordance with the instructions of button 2101. The case registration has the petition information "<Petition Name> X Company Nonpayment <Petition Amount> 100,000 <Main Payment Due Date> 2024 / 8 / 1 <Petitioner Identifier> C01 ... <Name of Respondent> X Company ... <Respondent's Phone Number> 090-1234-5566 <Respondent's Email Address> yy@x.jp". Next, the petitioner transmission unit 24 transmits the case registration to the negotiation support device 1. Note that the petitioner identifier "C01" is, for example, stored in the petitioner storage unit 21.

[0459] Next, the claim acquisition unit 121 of the negotiation support device 1 receives the case registration from the claimant terminal 2. The claim acquisition unit 121 obtains the current date and time, the time of filing, "2024 / 8 / 11 09:00", from a clock (not shown). The claim acquisition unit 121 constructs the claim information to be stored as "<Case Name>X Company Nonpayment <Claim Amount>100,000 <Main Payment Due Date>2024 / 8 / 1 <Claimant Identifier>C01 ··· <Name of Counterparty>X Company ··· <Counterparty Email Address>yy@x.jp <Time of Filing>2024 / 8 / 11 15:35". Next, the claim storage unit 131 obtains a unique claim identifier "1". Next, the claim storage unit 131 stores the constructed claim information in the claim management table (Figure 18) in pairs with the claim identifier "1". This claim information is the claim information with "ID=1" in Figure 18. The update unit 132 also stores the phase identifier "1" in the record with "ID=1" in Figure 18.

[0460] Next, the petition notification unit 141 of the negotiation support device 1 configures the notification information as follows. First, the petition notification unit 141 performs the notification information configuration process described using the flowchart in Figure 10, and configures the notification information in Figure 32. The petition notification unit 141 then configures the notification information by substituting the information corresponding to each underlined part into the notification information template in Figure 32, where the underlined parts are variables. At this point, the petition notification unit 141 may obtain the respondent's contact information "yy@x.jp" from the petition information and send the notification information (Figure 32) to the contact person identified by that contact information.

[0461] Next, the decision unit 137 performs the decision processing described using the flowchart in Figure 7. Specifically, the decision unit 137 obtains the phase identifier "Application Filed (1)" and determines that the phase identifier satisfies the agent cooperation condition for "ID=1" in Figure 29. In other words, the decision unit 137 decides to "perform dialogue processing".

[0462] Next, the Negotiation Support Unit 134 performs the dialogue processing described using the flowchart in Figure 8. First, the agent means 1341 obtains a prompt ("ID=2") that corresponds to the agent linkage condition (phase identifier = filed) from the agent linkage condition management table in Figure 29. The agent means 1341 also obtains the claim information, which is the information to be given to the generating AI. Next, the agent means 1341 assigns the claim information "<Claim Case Name>X Company Nonpayment <Claim Amount>100,000 <Main Payment Due Date>2024 / 8 / 1 <Claimant Identifier>C01 ··· <Counterparty Name>X Company ··· <Counterparty Phone Number>090-1234-5566 <Counterparty Email Address>yy@x.jp" to the variable <Respondent.Contact Information> of the obtained prompt, and assigns "090-1234-5566" to the variable <Respondent.Contact Information>. Next, the agent means 1341 provides the generating AI with prompts in which information has been substituted into each variable.

[0463] The generating AI then dials the respondent's phone number, "090-1234-5566," and the AI ​​agent based on the generating AI converses with Yamada Y. of Company X, the respondent, by phone, following the prompts in Figure 21. The processing of the generating AI can also be considered as the processing of agent means 1341.

[0464] The respondent then states over the phone that they "can pay in installments" and that the number of installments should be "20". The AI ​​agent then obtains the installment amount of "5,000 yen" from the billing amount of "100,000 yen" and the number of installments of "20". The AI ​​agent also returns the value "2" corresponding to the statement "I can pay in installments".

[0465] Next, the device linkage means 13411 obtains the phase identifier "claim filed" and the return value "2" from the AI ​​agent. Then, the device linkage means 13411 determines that the device linkage condition, which is the input condition "information indicating payment in installments" corresponding to the phase identifier condition "claim filed" and the return value "2", is satisfied. Alternatively, the device linkage means 13411 may, for example, perform speech recognition on the respondent's voice information, obtain a string, obtain the similarity between the vectorized string information and the vectorized input condition information, and determine that the input condition, which is threshold information, is satisfied.

[0466] Next, the device linkage means 13411 obtains the processing identifier "Send_message($petitioner.contact information,$acquired information)" which is paired with the phase identifier condition "petition filed" and the input condition "information that payment will be made in installments" from the device linkage condition management table (Figure 30). Next, the device linkage means 13411 assigns the petitioner's contact information (ta@c.com) to the variable "$petitioner.contact information" of the processing identifier, and assigns the information that "payment can be made in installments", the number of installments "20", and the installment amount "5000 yen" to the variable "$acquired information". Next, the device linkage means 13411 executes "Send_message(ta@c.com,“Payment can be made in installments <number of installments>20 <installment amount>5000 yen”)". As a result, the information “Payment can be made in installments <number of installments>20 <installment amount>5000 yen” is sent to Tanaka A-ko via email.

[0467] Let's say Tanaka A-ko, after viewing information regarding installment payments from the respondent, logs into negotiation support device 1 and enters her desired installment terms. In other words, for example, let's say Tanaka A-ko's petitioner terminal 2 accesses negotiation support device 1, and on the screen shown in Figure 23 displayed on petitioner terminal 2, she clicks the "Propose a change in installment amount" button 2301. Then, petitioner terminal 2 receives this instruction and sends a screen output instruction for the "Propose a change in installment amount" button 2301 to negotiation support device 1.

[0468] Next, the instruction receiving unit 124 of the negotiation support device 1 receives the screen output instruction. Next, the information acquisition unit 139 acquires the screen information corresponding to the screen output instruction. Next, the information transmission unit 147 transmits the screen information to Tanaka A's petitioner terminal 2.

[0469] Next, Tanaka A's petitioner terminal 2 receives the screen information and outputs the installment amount change screen shown in Figure 24. Then, Tanaka A enters the proposed new installment amount "20,000 yen" on the screen in Figure 24 (Figure 24, 2401).

[0470] Next, the petitioner's terminal 2 receives the proposed new installment amount of "20,000 yen" and transmits the negotiation information, including the said installment amount of "20,000 yen," to the negotiation support device 1, paired with the petition identifier "1."

[0471] Next, the negotiation receiving unit 123 of the negotiation support device 1 receives negotiation information including the installment amount "20,000 yen" paired with the application identifier "1".

[0472] Next, the proposal content acquisition unit 136 determines that the received negotiation information meets the proposal content creation condition "includes installment payments".

[0473] Next, the proposal content acquisition unit 136 performs the proposal content configuration process described using the flowchart in Figure 11. Specifically, the proposal content acquisition unit 136 acquires a proposal content template that corresponds to the proposal content creation conditions from the storage unit 11, and substitutes the installment amount "20,000 yen," the petitioner's name, the respondent's name, and the petition amount that correspond to the petition identifier "1" into the proposal content template to configure the proposal content. Next, the transmission unit 14 transmits the proposal content to Tanaka A's petitioner terminal 2.

[0474] Next, the negotiation storage unit 135 obtains the claim identifier "1" which is paired with the negotiation information. The negotiation storage unit 135 also obtains the time information "2024 / 8 / 11 16:34" from a clock (not shown). The negotiation storage unit 135 also obtains the claim amount "100,000 yen" and the installment amount "20,000 yen" which are paired with the claim identifier "1", and obtains the payment start date, payment end date, and payment amount for each month (obtains the payment schedule), and stores them paired with the claim identifier "1".

[0475] Next, the proposal content acquisition unit 136 acquires the previously acquired proposal content. Then, the proposal content acquisition unit 136 associates the proposal content with the negotiation information (file "P2") and stores it in the application management table (Figure 18).

[0476] Next, the processing unit 13 determines that the transmission condition for sending specific screen information, "the proposed content has been configured," is met. Next, the information acquisition unit 139 configures screen information including the negotiation information and link information to the proposed content. Next, the information transmission unit 147 transmits the screen information to Yamada Y's respondent terminal 3. The information transmission unit 147 may also transmit a link (URL) to the screen information to Yamada Y's respondent terminal 3.

[0477] Next, the respondent terminal 3 receives the screen information and outputs the screen shown in Figure 25. Then, Yamada Y. instructs the user to press the "Accept the revised installment amount" button 2501 in Figure 25. The respondent terminal 3 then accepts the instruction to press the button 2501. Next, the respondent terminal 3 obtains the information "<Negotiation Result> Agreement" associated with the button 2501, constructs negotiation information "<Petition Identifier>1 <Negotiation Result> Agreement <Installment Amount>20000 <Respondent Identifier>Company X" containing this information, and transmits it to the negotiation support device 1.

[0478] Next, the negotiation receiving unit 123 of the negotiation support device 1 receives the negotiation information. The negotiation storage unit 135 obtains the claim identifier "1". The negotiation storage unit 135 obtains the current time, the time of agreement "2024 / 8 / 11 16:39", from a clock (not shown). Next, the negotiation storage unit 135 stores the negotiation information "agreement" and the time of agreement in association with the claim identifier "1".

[0479] Next, the judgment unit 137 determines that the received negotiation information matches the agreement condition "The negotiation information includes '<negotiation result> agreement'." Then, the agreement document generation unit 138 generates the agreement document "A1" using the agreement document generation process explained using the flowchart in Figure 12. The agreement document generation unit 138 then stores the agreement document "A1" in the application management table (Figure 18) paired with the application identifier "1."

[0480] Next, the information acquisition unit 139, in response to the respondent's (Yoko Yamada) button 2501 (Figure 25), acquires screen information corresponding to the negotiation information "<Petition Identifier>1 <Negotiation Result>Agreement <Installment Amount>20000 <Respondent Identifier>Company X", which is to be sent to the petitioner. Next, the information transmission unit 147 transmits this screen information to the petitioner's terminal 2.

[0481] Furthermore, the information acquisition unit 139 acquires screen information that allows the respondent to determine the petition processing identifier, in response to the respondent's button 2501 (Figure 25). Next, the information transmission unit 147 transmits this screen information to the respondent's terminal 3.

[0482] Next, the petitioner's terminal 2 receives the screen information transmitted by the information transmission unit 147 and outputs the screen shown in Figure 26. The screen in Figure 26 is an example of a screen showing the process of filing the petition, confirming the case, and negotiations leading to an agreement.

[0483] Furthermore, the respondent terminal 3 receives screen information for determining the petition processing identifier and outputs a screen for determining the petition processing identifier. An example of such a screen is shown in Figure 27.

[0484] Next, it is stated that Yamada Y. instructed the respondent terminal 3 to click the "Pay by bank transfer" button 2701 on the screen shown in Figure 27.

[0485] Next, the respondent terminal 3 receives the instruction and obtains the application processing identifier "B". It should be noted that application processing identifier "A" represents "payment by credit card" and "B" represents "payment by bank transfer". Next, the respondent terminal 3 transmits the application processing identifier "B" to the negotiation support device 1, paired with application identifier "1".

[0486] Next, the negotiation support device 1 receives the application processing identifier "B," etc., and stores the application processing identifier "B" in the application management table (Figure 18) in conjunction with the application identifier "1." As a result of this process, the application processing identifier "3,B" is stored in the record "ID=1" in the application management table.

[0487] Furthermore, the update unit 132 of the negotiation support device 1 updates the phase identifier, which is paired with the claim identifier "1", to "5".

[0488] Next, the information acquisition unit 139 acquires the petitioner identifier "C01," which is paired with the petitioner identifier "1," from the petition management table (Figure 18). Next, the information acquisition unit 139 acquires the deposit destination information "Bank Account A," which is paired with the petitioner identifier "C01." Next, the information acquisition unit 139 constructs screen information including the said deposit destination information "Bank Account A." Next, the information transmission unit 147 transmits the said screen information to the respondent terminal 3.

[0489] Next, the respondent's terminal 3 receives the screen information and outputs the screen shown in Figure 28. In Figure 28, 2801 is the information for "Bank Account A".

[0490] As described above, according to this embodiment, the negotiation support device 1, which assists in negotiations between the petitioner and the respondent regarding a petition, can appropriately utilize an AI agent.

[0491] Furthermore, according to this embodiment, appropriate processing can be performed on the negotiation support device at an appropriate time when the respondent's statement satisfies the device linkage conditions.

[0492] Furthermore, according to this embodiment, if the respondent tells the AI ​​agent that they intend to conduct negotiations using the negotiation support device 1, the process can transition to negotiations using the negotiation support device 1.

[0493] Furthermore, according to this embodiment, negotiations with the respondent using an AI agent can be conducted at an appropriate time when the petitioner's input satisfies the agent collaboration conditions.

[0494] Furthermore, according to this embodiment, negotiations with the respondent using an AI agent can be conducted at an appropriate time after the petitioner has entered the petition information.

[0495] Furthermore, according to this embodiment, the AI ​​agent and the respondent can negotiate by telephone.

[0496] Furthermore, according to this embodiment, if the respondent does not answer the call from the AI ​​agent, the AI ​​agent can call the respondent at the following timing.

[0497] Furthermore, according to this embodiment, it is possible to appropriately answer questions from the petitioner or respondent regarding the negotiation support device 1.

[0498] The processing in this embodiment may be implemented by software. This software may be distributed by software download or the like. Alternatively, this software may be recorded on a recording medium such as a CD-ROM and distributed. This also applies to other embodiments in this specification. The software that implements the negotiation support device 1 in this embodiment is the following program. In other words, this program is a program that causes the computer to function as a claim acquisition unit that acquires claim information having claim content information relating to the content of the claim, a claim storage unit that stores the claim information, and a negotiation support unit that supports negotiations between the claimant who made the claim and the respondent against whom the claim is made, and the negotiation support unit is equipped with agent means that performs dialogue processing, which is the processing of interacting with the respondent on behalf of the claimant, in at least a part of the phases from the start to the end of negotiations between the claimant and the respondent.

[0499] (Embodiment 2) In this embodiment, we describe a negotiation support device in which an AI agent acting as a proxy to assist in negotiations on behalf of the parties involved, acquiring and outputting information during at least some phases of the process, from the start of negotiations between the claimant and the respondent to the end of negotiations or the completion of the fulfillment of the agreed-upon terms.

[0500] Furthermore, role-acting AI agents are AI agents that act on behalf of the roles of the parties involved. Role-acting AI agents include petitioner AI agents, which negotiate on behalf of the petitioner; respondent AI agents, which negotiate on behalf of the respondent; and third-party AI agents, which support negotiations on behalf of third parties. Third-party AI agents include, for example, mediator AI agents, which represent mediators; arbitrator AI agents, which represent arbitrators; supporter AI agents, which represent supporters; or secretariat AI agents, which represent secretariat staff.

[0501] This embodiment describes a negotiation support device in which a support AI agent assists the parties involved in at least some phases of negotiations between the petitioner and the respondent, from the start of negotiations to the end of negotiations or the completion of the agreement, acquires information and outputs it to the party being supported. The support AI agent is, for example, a petitioner support AI agent that assists the petitioner, a respondent support AI agent that assists the respondent, or a third-party support AI agent that assists a third party. The third-party support AI agent is a mediator support AI agent that assists the mediator, an arbitrator support AI agent that assists the arbitrator, a supporter support AI agent that assists the supporter, or an office support AI agent that assists the office.

[0502] In this embodiment, a negotiation support device that causes an AI agent to generate and output one or more settlement proposals will be described.

[0503] In this embodiment, we will describe a negotiation support device that, when the conditions for creating a settlement proposal are met, causes an AI agent to generate and output a settlement proposal.

[0504] In this embodiment, a negotiation support device in which an AI agent operates when the agent collaboration conditions are met will be described.

[0505] This embodiment describes a negotiation support device that includes a neutrality judgment AI agent that performs a neutrality assessment regarding negotiations, obtains the neutrality assessment results, and outputs them.

[0506] In this embodiment, a negotiation support device that outputs a neutrality evaluation result when the neutrality output conditions are met will be described.

[0507] Figure 33 is a conceptual diagram of information system B in this embodiment. Information system B comprises a negotiation support device 5, one or more petitioner terminals 2, one or more respondent terminals 3, one or more third-party terminals 6, and one or more generation AI devices 4.

[0508] Negotiation support device 5 is a device that supports negotiations between the claimant and the respondent. Negotiation support device 5 is a device on which a role-acting AI agent and / or a support AI agent functions. Negotiation support device 5 is a device that provides information processing functions in, for example, a cloud environment, ASP environment, SaaS environment, etc., and its implementation form is not limited. Third-party terminal 6 is a terminal used by a third party. Third-party terminal 6 is, for example, a so-called personal computer, smartphone, or tablet terminal, but the type is not limited. It is preferable that negotiation support device 5 has all or part of the configuration and functions of negotiation support device 1.

[0509] The negotiation support device 5 and one or more petitioner terminals 2, the negotiation support device 5 and one or more respondent terminals 3, the negotiation support device 5 and one or more third-party terminals 6, and the negotiation support device 5 and one or more generating AI devices 4 can communicate with each other via a network such as the Internet.

[0510] The negotiation support device 5 and one or more petitioner terminals 2, the negotiation support device 5 and one or more respondent terminals 3, and the negotiation support device 5 and one or more third-party terminals 6 communicate using one or more means from among chat, email, SMS, voice, telephone, etc. In other words, the means of communication between two or more devices is not limited. For example, when an participant and an AI agent interact, one or more means from among chat, email, SMS, voice, telephone, etc. will be used.

[0511] Figure 34 is a block diagram of information system B in this embodiment. Figure 35 is a block diagram of negotiation support device 5.

[0512] The negotiation support device 5 comprises a storage unit 51, a receiving unit 52, a processing unit 53, and a transmission unit 54. The storage unit 51 comprises a claim management unit 511, a condition management unit 512, a response information storage unit 113, and a prompt management unit 514. The receiving unit 52 comprises a claim acquisition unit 121, a confirmation receiving unit 122, a negotiation receiving unit 123, and an instruction receiving unit 524. The processing unit 53 comprises a claim storage unit 131, an update unit 132, a confirmation storage unit 133, a negotiation support unit 534, a negotiation storage unit 135, a proposal content acquisition unit 136, a judgment unit 537, an agreement document generation unit 138, and an information acquisition unit 139. The negotiation support unit 534 comprises an agent means 5341. The agent means 5341 includes a device linkage means 13411, a learning means 13412, a response means 13413, a neutrality evaluation means 53411, and a settlement proposal means 53412. The transmission unit 54 includes a claim notification unit 141, a screen transmission unit 142, a proposal content transmission unit 143, a negotiation information output unit 144, a necessary information transmission unit 145, an agreement document output unit 146, and an information transmission unit 147.

[0513] The petitioner's terminal 2 can communicate with the negotiation support device 5, the respondent's terminal 3, and the third-party terminal 6, for example, via chat, phone, SMS, email, etc.

[0514] The respondent's terminal 3 can communicate with the negotiation support device 5, the petitioner's terminal 2, and the third-party terminal 6, for example, via chat, phone, SMS, email, etc.

[0515] The third-party terminal 6 comprises a third-party storage unit 61, a third-party receiving unit 62, a third-party processing unit 63, a third-party transmission unit 64, a third-party receiving unit 65, and a third-party output unit 66. The third-party terminal 6 can communicate with the negotiation support device 5, the petitioner terminal 2, the respondent terminal 3, and other third-party terminals 6, for example, via chat, telephone, SMS, email, etc.

[0516] The storage unit 51, which constitutes the negotiation support device 5, stores various types of information. These types of information include, for example, application status information, various screen information, various conditions, response information, various prompts, various templates, a set of static information, and a set of dynamic information. The set of static information consists of one or more types of static information. The set of dynamic information consists of one or more types of dynamic information.

[0517] Static information includes, for example, laws, articles, amendment history, court precedents, case commentaries, legal literature, articles, academic theories, forms, procedural information, industry information, industry-specific legal risks, industry practices, administrative documents, comparative legal information, foreign legal systems, or legal glossaries.

[0518] Dynamic information includes, for example, application information, consultation details, party input, chat history, audio transcripts, phone logs, submitted documents, evidence, party comments, mediator comments, secretariat comments, AI proposal history, mediator selection / modification / rejection logs, settlement proposals, settlement / rejection results, performance status, recovery results, participation rate, response rate, and time to resolution. Party input includes, for example, negotiation information or dialogue information described later. Chat history, audio transcripts, and phone logs are examples of dialogue information. Party comments, mediator comments, and secretariat comments are negotiation information or dialogue information. AI proposal history is generated information. Settlement / rejection results are information included in the application status information. Generated information is information output by any of the AI ​​agents.

[0519] The static and dynamic information of the storage unit 51 may be obtained from an external device (e.g., a server such as a cloud server) not shown in the diagram, and may be information that exists at least temporarily.

[0520] The application management unit 511 stores one or more application status information. The application status information preferably includes respondent information. The respondent information preferably includes the respondent's telephone number. In addition to the application status information in the application management unit 111, the application status information preferably includes agent management information and dialogue information. Dialogue information refers to information indicating the content of conversations between the participants and the AI ​​agent. Dialogue information is generated information output from the AI ​​agent or information entered by the participants. Dialogue information may also be considered to include negotiation information.

[0521] Agent management information is information used to identify functioning AI agents. Agent management information may include, for example, one or more agent type identifiers. If the agent management information is NULL, it indicates that no AI agent is operational in the negotiations for that claim. An agent type identifier is information that identifies the type of AI agent.

[0522] Agent type identifiers include, for example, the petitioner AI agent (e.g., "1"), the respondent AI agent (e.g., "2"), the third-party AI agent, the petitioner support AI agent (e.g., "3"), the respondent support AI agent (e.g., "4"), and the third-party support AI agent. Third-party AI agents include, for example, the mediator AI agent (e.g., "5"), the arbitrator AI agent (e.g., "6"), the supporter AI agent (e.g., "7"), or the secretariat AI agent (e.g., "8"). Third-party support AI agents include, for example, the mediator support AI agent (e.g., "9"), the arbitrator support AI agent (e.g., "10"), the supporter support AI agent (e.g., "11"), or the secretariat support AI agent (e.g., "12").

[0523] The condition management unit 512 stores one or more conditions. These conditions may include, for example, one or more agent cooperation conditions, one or more settlement proposal creation conditions, one or more neutrality evaluation conditions, one or more device cooperation conditions, one or more proposal content creation conditions, or one or more agreement conditions. It goes without saying that these various conditions may also be embedded within the program.

[0524] Each of the one or more conditions may be associated with a claim identifier. Each of the one or more conditions may be associated with a user identifier. Each of the one or more conditions may be associated with one or more user attribute values. In other words, the various conditions may differ for each claim, for each user (e.g., each claimant), and for each condition based on user attribute values.

[0525] The agent collaboration conditions, in this context, include, for example, the petitioner's agent collaboration conditions, the respondent's agent collaboration conditions, the third-party agent collaboration conditions, the petitioner's supporting agent collaboration conditions, the respondent's supporting agent collaboration conditions, and the third-party supporting agent collaboration conditions.

[0526] Agent collaboration conditions are the conditions for performing proxy processing by an AI agent, or for performing support processing by an AI agent. In other words, agent collaboration conditions are the conditions for performing proxy processing or support processing by agent means 5341. Agent collaboration conditions include, for example, conditions based on application information and conditions based on the negotiation status.

[0527] Conditions based on the claim information include, for example, conditions regarding the claim amount included in the claim information. Conditions regarding the claim amount include the claim amount being less than or equal to a threshold.

[0528] Conditions based on the negotiation status include, for example, conditions based on the current phase (e.g., at the time of filing, at the time of notification to the respondent, at the time of agreement), conditions based on the elapsed time from a specific point in time (e.g., at the time of filing, at the time of notification to the respondent, at the time of agreement), and conditions based on negotiation information. Conditions based on negotiation information include, for example, conditions based on the number of times negotiation information has been sent and received, and conditions based on the content of the negotiation information. A condition based on the number of times negotiation information has been sent and received is, for example, that the number of times negotiation information has been received is equal to or greater than a threshold. A condition based on the content of negotiation information is, for example, that the classification result (class) of the received negotiation information was a specific class (e.g., "negative," "emotional," "rejected"). Note that the process for obtaining the classification result of negotiation information is publicly known technology.

[0529] Agent collaboration conditions are, for example, conditions based on the current phase. Agent collaboration conditions are, for example, conditions based on applicant input. Applicant input refers to the input from the applicant. Applicant input includes, for example, application information, negotiation information, and dialogue information.

[0530] Agent collaboration conditions are, for example, conditions based on respondent input. Respondent input refers to the information entered by the respondent. Respondent input includes, for example, confirmation information, negotiation information, and dialogue information.

[0531] Agent collaboration conditions are, for example, conditions based on third-party input. Third-party input refers to input from a third party. Third-party input may include negotiation information or dialogue information.

[0532] Agent linkage conditions are, for example, conditions based on the current phase and applicant input. Agent linkage conditions are, for example, conditions based on the current phase and respondent input. Agent linkage conditions are, for example, conditions based on the current phase and third-party input. Agent linkage conditions include, for example, conditions for phase identifiers and conditions for applicant input, respondent input, or third-party input. Agent linkage conditions are, for example, conditions regarding information on the status of compliance with the agreement.

[0533] Agent integration conditions include, for example, the receipt of a company notification from the claimant. Agent integration conditions include, for example, the receipt of claim information from the claimant. Agent integration conditions include, for example, the information on the status of performance of the agreement is "information indicating that the agreement has not been performed." Claimant input refers to input from the claimant. Respondent input refers to input from the respondent. Furthermore, the functions of the AI ​​agent are, for example, functions realized by the generative AI.

[0534] The petitioner agent integration conditions are the conditions required for the petitioner AI agent to function. The petitioner AI agent is an AI agent that interacts on behalf of the petitioner with the respondent, a third party, the respondent's AI agent, or one or more third-party AI agents. The petitioner agent integration conditions include, for example, that the agent has received instructions from the petitioner, that it has received negotiation information sent by the respondent, and that the class of the negotiation information sent by the respondent is a specific class (e.g., "emotional" or "negative").

[0535] The Respondent AI Agent Integration Conditions are the conditions required for the Respondent AI Agent to function. The Respondent AI Agent is an AI agent that interacts on behalf of the Respondent with the Petitioner, a third party, the Petitioner's AI Agent, or one or more third-party AI Agents. Respondent Agent Integration Conditions include, for example, the receipt of instructions from the Respondent, the receipt of negotiation information sent by the Petitioner, and the fact that the class of the negotiation information sent by the Petitioner was a specific class (e.g., "emotional" or "negative").

[0536] Third-party AI agent integration conditions are the conditions required for a third-party agent to function. A third-party AI agent is an AI agent that interacts with the claimant or respondent or the claimant's AI agent or the respondent's AI agent or other third-party AI agents on behalf of a third party. Third-party agent integration conditions include, for example, the receipt of instructions from the third party, the receipt of negotiation information sent by the claimant or respondent, and the fact that the class of the negotiation information sent by the claimant or respondent was a specific class (e.g., "negative," "emotional," "rejection").

[0537] The conditions for collaborating with the petitioner support agent are the conditions required for the petitioner support AI agent to function. These conditions include, for example, that a support instruction has been received from the petitioner, that negotiation information has been received from the respondent, and that the class of the negotiation information received from the respondent is a specific class (e.g., "negative," "emotional," "rejection").

[0538] The conditions for the respondent support agent to function are the conditions required for the respondent support AI agent to operate. These conditions include, for example, that a support instruction has been received from the respondent, that negotiation information has been received from the petitioner, and that the class of the negotiation information received from the petitioner is a specific class (e.g., "negative," "emotional," "rejection").

[0539] Third-party support agent integration conditions are the conditions required for a third-party support AI agent to function. A third-party support AI agent is an AI agent that provides support to third parties. Third-party support agent integration conditions include, for example, that support instructions have been received from a third party, that negotiation information has been received from the claimant or respondent, and that the class of the negotiation information received from the claimant or respondent was a specific class (e.g., "negative," "emotional," "rejection").

[0540] The conditions for drafting a settlement proposal are the conditions required to create a settlement proposal. These conditions may include, for example, conditions based on the status of negotiations, or the receipt of instructions (settlement proposal drafting instructions) from the petitioner, respondent, or a third party.

[0541] Neutrality assessment conditions are the conditions used to assess neutrality. These conditions may include, for example, receiving negotiation information transmitted by a third party, obtaining generated information output by a third-party AI agent, conditions based on the negotiation situation, or a combination of two or more of these conditions.

[0542] The prompt management unit 514 stores various prompts. Prompts can also be called input information. Input information includes, for example, instructions, commands, questions, requests, and conditions given to the generating AI. Various prompts include, for example, system prompts for configuring various AI agents, such as AI agent configuration prompts, neutrality judgment AI agent configuration prompts, settlement proposal generation AI agent configuration prompts, negotiator prompts, inquiry prompts, agreement judgment prompts, or inquiry judgment prompts. Negotiator prompts include, for example, post-company notification prompts or post-claim prompts. Note that two or more prompts may be included in a single prompt.

[0543] Examples of AI agent configuration prompts include the petitioner AI agent configuration prompt, the respondent AI agent configuration prompt, the third-party AI agent configuration prompt, the petitioner support AI agent configuration prompt, the respondent support AI agent configuration prompt, and the third-party support AI agent configuration prompt.

[0544] The Complainant AI Agent Configuration Prompt is a system prompt for configuring the Complainant AI Agent. For example, the Complainant AI Agent Configuration Prompt might say, "You are a dispute resolution expert. You will negotiate on behalf of the dispute complainant..."

[0545] The Respondent AI Agent Configuration Prompt is a system prompt for configuring the Respondent AI Agent. For example, the Respondent AI Agent Configuration Prompt might say, "You are a dispute resolution expert. You will negotiate on behalf of the respondent in the dispute..."

[0546] A third-party AI agent configuration prompt is a system prompt for configuring a third-party AI agent. For example, a third-party AI agent configuration prompt might say, "You are a dispute resolution expert. You will assist in negotiations on behalf of a third party assisting in dispute resolution..." The third-party AI agent configuration prompt may also include the content of the neutrality judgment AI agent configuration prompt, which will be discussed later.

[0547] The Complaint Support AI Agent Configuration Prompt is a system prompt for configuring the Complaint Support AI Agent. For example, the Complaint Support AI Agent Configuration Prompt might read: "You are a dispute resolution expert. You will play a role in assisting the complainant in resolving the dispute..."

[0548] The Respondent Support AI Agent Configuration Prompt is a system prompt for configuring the Respondent Support AI Agent. For example, the Respondent Support AI Agent Configuration Prompt might say, "You are a dispute resolution expert. You will play a role in supporting the respondent in resolving the dispute..."

[0549] The Third-Party Support AI Agent Configuration Prompt is a system prompt for configuring a Third-Party Support AI Agent. For example, the Third-Party Support AI Agent Configuration Prompt might say, "You are a dispute resolution expert. You will play a role in supporting third parties in resolving disputes..." The Third-Party Support AI Agent Configuration Prompt may also include the content of the Neutrality Judgment AI Agent Configuration Prompt, which will be discussed later.

[0550] The Neutrality Judgment AI Agent Configuration Prompt is a system prompt for configuring a neutrality judgment AI agent. A neutrality judgment AI agent is an AI agent that performs a neutrality assessment regarding negotiations, obtains the neutrality assessment result, and outputs it. The neutrality judgment AI agent configuration prompt may look like this: "You are a professional who assists in resolving disputes. Based on the claim information, negotiation information submitted by the claimant, respondent, or a third party, and generated information output by other AI agents, perform a neutrality assessment of this negotiation, obtain the neutrality assessment result, and output it. The neutrality assessment is on a 5-point scale from "1" to "5," where "1" indicates a bias towards the claimant, "3" indicates complete neutrality, and "5" indicates a bias towards the respondent."

[0551] The Settlement Proposal Generation AI Agent Configuration Prompt is a system prompt for configuring the Settlement Proposal Generation AI Agent. The Settlement Proposal Generation AI Agent is an AI agent that generates and outputs settlement proposals for negotiations. The Settlement Proposal Generation AI Agent Configuration Prompt is a prompt that includes instructions for creating a settlement proposal for a dispute. For example, the Settlement Proposal Generation AI Agent Configuration Prompt might say: "You are an expert in creating settlement proposals for disputes. When creating a settlement proposal, please refer to the claim information, past negotiation information, and dialogue information. Also, when creating a settlement proposal, please generate multiple candidate solutions that consider not only the claims and wishes of both parties, but also the following: • Legal prospects (likelihood of winning or losing if the case goes to court) • Realistic ability to pay • Possibility of recovery • Cost of going to court • Time cost • Attorney fees • Procedural costs • Statistics of similar past cases • Likelihood of settlement • Likelihood of performance • Likelihood of acceptance by the parties. Examples of settlement proposals include lump-sum payment, installment payments, partial reduction, extension of deadline, apology, prevention of recurrence, contract modification, termination of relationship, etc."

[0552] The receiving unit 52 receives various instructions and information. These instructions and information include, for example, input from the petitioner, input from the respondent, input from a third party, proxy instructions described later, support instructions described later, and dialogue information. The receiving unit 52 may receive information or instructions by means of input to a chat, by telephone, by SMS, or by email, for example.

[0553] The receiving unit 52 receives applicant input from the applicant terminal 2. Applicant input includes, for example, company notices, application information, and negotiation information.

[0554] The receiving unit 52 receives the respondent's input from the respondent's terminal 3. The respondent's input may include, for example, confirmation information and negotiation information.

[0555] The receiving unit 52 receives third-party input from the third-party terminal 6. Third-party input may include, for example, confirmation information or negotiation information.

[0556] The instruction receiving unit 524 receives various instructions. These instructions include, for example, instructions for acting on behalf of others or instructions for providing support.

[0557] An authorization to act on behalf of another party is an instruction sent by an involved party to have an AI agent act on their behalf. Examples of authorization to act on behalf of another party include an authorization to have the petitioner's AI agent act on behalf of the petitioner, an authorization to have the respondent's AI agent act on behalf of the respondent, and an authorization to have a third-party AI agent act on behalf of a third party.

[0558] Support instructions are instructions sent by an involved party, instructing an AI agent to provide support to that party. Examples of support instructions include a petitioner support instruction, which instructs the petitioner support AI agent to support the petitioner; a respondent support instruction, which instructs the respondent AI agent to support the respondent; and a third-party support instruction, which instructs a third-party AI agent to support a third party.

[0559] The processing unit 53 performs various processes. These processes include, for example, those performed by the application storage unit 131, the update unit 132, the confirmation storage unit 133, the negotiation support unit 534, the negotiation storage unit 135, the proposal content acquisition unit 136, the judgment unit 537, the agreement document generation unit 138, and the information acquisition unit 139.

[0560] Negotiation Support Unit 534 assists in negotiations between the claimant and the respondent. Negotiation Support Unit 534 assists in negotiations by, for example, one or more AI agents from among one or more role-playing AI agents and one or more supporting AI agents. Assisting in negotiations typically means assisting in the agreement or performance of the negotiations.

[0561] Negotiation Support Unit 534 conducts negotiations with the respondent, for example, using the respondent's AI agent, which is a virtual agent of the claimant. Negotiation Support Unit 534 conducts negotiations with the claimant, for example, using the respondent's AI agent, which is a virtual agent of the respondent. Negotiation Support Unit 534 provides support for negotiations, for example, using a third-party AI agent, which is a virtual agent of a third party.

[0562] Negotiation Support Unit 534 assists the petitioner, for example, with a petitioner support AI agent. Negotiation Support Unit 534 assists the respondent, for example, with a respondent support AI agent. Negotiation Support Unit 534 assists third parties, for example, with a third-party support AI agent.

[0563] The agent means 5341 implements one or more AI agents. The roles of the two or more AI agents described below may be implemented by a single AI agent. Also, the role of a single AI agent described below may be implemented by two or more AI agents.

[0564] The agent means 5341 implements, for example, one or more role-acting AI agents. The agent means 5341 implements, for example, one or more support AI agents.

[0565] A role-acting AI agent is an AI agent that negotiates on behalf of or assists in negotiations on behalf of the claimant, respondent, or a third party, and performs proxy processing, which is the process of communicating with other parties on behalf of the claimant, respondent, or a third party.

[0566] A support AI agent is an AI agent that performs support processing to obtain information to be output to the claimant, respondent, or third party in order to support the claimant, respondent, or third party in negotiations.

[0567] The agent means 5341, for example, provides the generating AI with one or more AI agent configuration prompts, thereby configuring one or more AI agents. The timing of the agent means 5341 configuring one or more AI agents is not specified. When the agent means 5341 has already configured one or more AI agents, the negotiation support device 5 may support negotiations regarding the claim. It is preferable for the agent means 5341 to perform proxy processing by the AI ​​agent or support processing by the AI ​​agent when the agent cooperation conditions are met.

[0568] Agent means 5341 searches a set of static information and / or a set of dynamic information using, for example, claim information, negotiation information received during negotiations, dialogue information indicating the content of conversations between participants or AI agents and other AI agents during negotiations, or generated information output by AI agents during negotiations, and obtains one or more pieces of information. Next, agent means 5341 passes, for example, the one or more pieces of information to one or more role-acting AI agents and / or one or more support AI agents. In this way, the role-acting AI agents and / or support AI agents can assist in negotiations more effectively. The set of static information and / or a set of dynamic information is stored, for example, in storage unit 51 or in storage units of one or more external devices (not shown).

[0569] Agent means 5341, for example, provides a petitioner AI agent configuration prompt to the generating AI, and configures the petitioner AI agent based on that prompt. Then, agent means 5341 performs petitioner proxy processing. Note that the petitioner AI agent may be either the petitioner AI agent or the petitioner's representative AI agent, or both. The petitioner's representative AI agent is the AI ​​agent of the petitioner's representative.

[0570] The petitioner's proxy processing refers to the processing of dialogue with the respondent, a third party, or an AI agent. For example, the petitioner's proxy processing includes the following: For example, the petitioner's proxy processing includes proxy processing that receives dialogue information from the respondent, a third party, or another AI agent, passes said dialogue information to the petitioner's AI agent, obtains a response from the petitioner's AI agent, and outputs dialogue information (which may also be called generated information) based on the response to the respondent, a third party, or another AI agent. The agent means 5341, for example, provides the petitioner's AI agent with petition information and processes it to interact with the respondent, a third party, or an AI agent on behalf of the petitioner. The agent means 5341, for example, obtains petitioner agent output (which may also be called generated information), which is the output from the petitioner's AI agent, and outputs said petitioner agent output to the respondent, a third party, or an AI agent. Agent means 5341 acquires, for example, dialogue information transmitted by the respondent or a third party, or generated information which is output from another AI agent, and provides the negotiation information or generated information to the claimant AI agent. Providing a prompt to the generating AI can be thought of as inputting the prompt to the generating AI, passing the prompt to the generating AI, inputting the prompt to the generating AI and executing it, etc. Configuring an AI agent can be thought of as generating an AI agent, etc.

[0571] In the applicant's proxy processing, agent means 5341 searches a set of static information and / or a set of dynamic information using, for example, application information, negotiation information received during negotiations, dialogue information indicating the content of conversations between the applicant's AI agent and the other AI agents during negotiations, or generated information output by other AI agents during negotiations, and obtains one or more pieces of information. Next, agent means 5341 passes, for example, the one or more pieces of information to the applicant's AI agent.

[0572] Agent means 5341 is preferable to perform applicant proxy processing when the conditions for cooperation with the applicant agent are met.

[0573] Agent means 5341 provides a respondent AI agent configuration prompt to the generating AI, and configures the respondent AI agent based on the prompt. The respondent AI agent may be either the respondent AI agent or the respondent's representative AI agent, or both. The respondent's representative AI agent is the AI ​​agent of the respondent's representative.

[0574] Agent means 5341 provides the respondent AI agent with the claim information and performs respondent proxy processing. Respondent proxy processing is processing to interact with the claimant, a third party, or other AI agents on behalf of the respondent. Respondent proxy processing includes, for example, receiving dialogue information from the claimant or a third party, or acquiring generated information which is output from another AI agent, passing the negotiation information or the generated information to the respondent AI agent, obtaining a response from the respondent AI agent, and outputting generated information based on the response to the claimant, a third party, or other AI agent. Agent means 5341, for example, acquires respondent agent output (generated information) which is output from the respondent AI agent, and outputs the respondent agent output to the claimant, a third party, or other AI agent. Agent means 5341, for example, acquires dialogue information transmitted from the claimant, a third party, or other AI agent, and provides the respondent AI agent with the dialogue information.

[0575] In the respondent's proxy processing, agent means 5341 searches a set of static information and / or a set of dynamic information using, for example, the claim information, or negotiation information received during negotiations, or dialogue information indicating the content of conversations between the participants or other AI agents and the respondent AI agent during negotiations, or generated information output by other AI agents during negotiations, and obtains one or more pieces of information. Next, agent means 5341 passes the one or more pieces of information to the respondent AI agent, for example.

[0576] Agent means 5341 is preferable to perform respondent proxy processing when the respondent agent cooperation conditions are met.

[0577] Agent means 5341 provides the generating AI with one or more different third-party AI agent configuration prompts, and configures one or more third-party AI agents based on these prompts. Agent means 5341 then performs third-party proxy processing. The third-party AI agent may be either a third-party AI agent or a third-party agent AI agent, or both. A third-party agent AI agent is an AI agent acting as an agent for a third party.

[0578] Third-party proxy processing refers to processing for interacting with the petitioner, respondent, or other AI agents on behalf of a third party. Third-party proxy processing includes, for example, proxy processing that receives dialogue information from the petitioner or respondent, passes said dialogue information to a third-party AI agent, obtains a response from the third-party AI agent, and outputs generated information based on the response to the petitioner, respondent, or other AI agents. Third-party proxy processing also includes, for example, providing petition information to one or more third-party AI agents and interacting with one or more of the petitioner, respondent, or other third parties or AI agents on behalf of one or more third parties. Agent means 5341, for example, obtains third-party agent output (generated information), which is the output from a third-party AI agent, and outputs said third-party agent output to the petitioner, respondent, or other third parties or other AI agents. Agent means 5341, for example, obtains dialogue information transmitted from the petitioner, respondent, or other third parties, or generated information, which is the output of other AI agents, and provides said negotiation information or said generated information to the third-party AI agent.

[0579] In third-party proxy processing, agent means 5341 searches a set of static information and / or a set of dynamic information using, for example, claim information, negotiation information received during negotiations, dialogue information indicating the content of conversations between the participants or other AI agents and the third-party AI agent during negotiations, or generated information output by other AI agents during negotiations, and obtains one or more pieces of information. Next, agent means 5341 passes the one or more pieces of information to the third-party AI agent, for example.

[0580] The agent means 5341 is preferable to perform third-party proxy processing when the conditions for third-party agent cooperation are met.

[0581] Agent means 5341 provides a petitioner support AI agent configuration prompt to the generating AI, and configures the petitioner support AI agent based on the prompt. Then, agent means 5341 performs petitioner support processing. The petitioner support AI agent may be either a petitioner support AI agent or a petitioner's representative support AI agent, or both. A petitioner's representative support AI agent is an AI agent that supports the petitioner's representative.

[0582] The petitioner support process is as follows, for example: Agent means 5341 provides, for example, petition information, obtains a response from the petitioner support AI agent, obtains petitioner support agent output (which may also be called generated information) based on the response, and outputs the petitioner support agent output to the petitioner. Agent means 5341 provides, for example, petitioner input (which may also be called dialogue information) to the petitioner support AI agent, obtains a response from the petitioner support AI agent, and transmits generated information based on the response to the petitioner. For example, by feeding back such generated information to the petitioner, the petitioner reviews the petitioner input and sends the updated petitioner input to the negotiation support device 5. Agent means 5341 provides, for example, respondent input to the petitioner support AI agent, obtains a response from the petitioner support AI agent, and transmits generated information based on the response to the petitioner. Agent means 5341, for example, provides a third-party input to a petitioner support AI agent, obtains a response from the petitioner support AI agent, and transmits generated information based on the response to the petitioner.

[0583] In the petitioner support process, the agent means 5341 searches a set of static information and / or a set of dynamic information using, for example, petition information, negotiation information received during negotiations, dialogue information indicating the content of conversations between the participants or other AI agents and the petitioner support AI agent during negotiations, or generated information output by other AI agents during negotiations, and obtains one or more pieces of information. Next, the agent means 5341 passes, for example, the one or more pieces of information to the petitioner support AI agent.

[0584] It is preferable for agent means 5341 to perform petitioner support processing when the conditions for cooperation with petitioner support agents are met.

[0585] Agent means 5341 provides a respondent support AI agent configuration prompt to the generating AI, and configures the respondent support AI agent based on the prompt. Agent means 5341 then performs respondent support processing. The respondent support AI agent may be either the respondent support AI agent or the respondent's representative support AI agent, or both. The respondent's representative support AI agent is an AI agent that supports the respondent's representative.

[0586] The respondent support process is as follows, for example: Agent means 5341 provides, for example, at least the claim information to the respondent support AI agent, obtains a response from the respondent support AI agent, and outputs generated information based on the response to the respondent. Agent means 5341 provides, for example, the respondent input to the respondent support AI agent, obtains a response from the respondent support AI agent, and transmits generated information based on the response to the respondent. Agent means 5341 provides, for example, the respondent input to the respondent support AI agent, obtains a response from the respondent support AI agent, and transmits generated information based on the response to the respondent. For example, by feeding back such generated information to the respondent, the respondent can review the respondent input and send the updated respondent input to the negotiation support device 5. Agent means 5341, for example, provides third-party input to a respondent support AI agent, obtains a response from the respondent support AI agent, and transmits generated information based on the response to the respondent.

[0587] In the respondent support process, the agent means 5341 searches a set of static information and / or a set of dynamic information using, for example, the claim information, or negotiation information received during negotiations, or dialogue information indicating the content of conversations between the participants or other AI agents and the respondent support AI agent during negotiations, or generated information output by other AI agents during negotiations, and obtains one or more pieces of information. Next, the agent means 5341 passes the one or more pieces of information to, for example, the respondent support AI agent.

[0588] Agent means 5341 is preferable to perform respondent support processing when the respondent support agent cooperation conditions are met.

[0589] Agent means 5341 provides a third-party support AI agent configuration prompt to the generating AI and configures a third-party support AI agent based on the prompt. Agent means 5341 then performs third-party support processing. The third-party support AI agent may be either a third-party support AI agent or a third-party agent support AI agent, or both. A third-party agent support AI agent is an AI agent that supports a third-party agent.

[0590] Third-party support processing is as follows, for example: Agent means 5341 provides at least claim information to the third-party support AI agent, obtains a response from the third-party support AI agent, and outputs generated information based on the response to the claimant. Agent means 5341 provides claimant input to the third-party support AI agent, obtains a response from the third-party support AI agent, and transmits generated information based on the response to the third party. Agent means 5341 provides respondent input to the third-party support AI agent, obtains a response from the third-party support AI agent, and transmits generated information based on the response to the third party. Agent means 5341 provides third-party input of a third party different from the said third party, obtains a response from the said third-party support AI agent, and transmits generated information based on the response to the said third party. Agent means 5341 provides third-party input of the said third party to the third-party support AI agent, obtains a response from the said third-party support AI agent, and transmits generated information based on the response to the said third party. For example, by providing feedback of such generated information to the third party, the third party can review the third-party input and send the updated third-party input to the negotiation support device 5.

[0591] In third-party support processing, agent means 5341 searches a set of static information and / or a set of dynamic information using, for example, claim information, negotiation information received during negotiations, dialogue information indicating the content of conversations between the participants or other AI agents and the third-party support AI agent during negotiations, or generated information output by other AI agents during negotiations, and obtains one or more pieces of information. Next, agent means 5341 passes the one or more pieces of information to, for example, the third-party support AI agent.

[0592] It is preferable for the agent means 5341 to perform third-party support processing when the conditions for collaboration with a third-party support agent are met.

[0593] The neutrality evaluation means 53411 provides a neutrality judgment AI agent configuration prompt to the generating AI, and configures a neutrality judgment AI agent based on the prompt. The neutrality evaluation means 53411 then provides one or more types of information to the neutrality judgment AI agent, obtains a neutrality evaluation result from the neutrality judgment AI agent, and outputs the neutrality evaluation result or information based on the neutrality evaluation result. The one or more types of information provided to the neutrality judgment AI agent include, for example, claim information, negotiation information or dialogue information entered by the claimant, negotiation information or dialogue information entered by the respondent, negotiation information or dialogue information entered by a third party, or generated information. The neutrality evaluation result is an evaluation result regarding neutrality in negotiations.

[0594] The neutrality evaluation means 53411 searches a set of static information and / or a set of dynamic information using, for example, claim information, or negotiation information received during negotiations, or dialogue information indicating the content of conversations between the participants or other AI agents and the neutrality judgment AI agent during negotiations, or generated information output by other AI agents during negotiations, and obtains one or more pieces of information. Next, the agent means 5341 passes, for example, the one or more pieces of information to the neutrality judgment AI agent.

[0595] The neutrality evaluation means 53411 preferably outputs the neutrality evaluation result or information based on the neutrality evaluation result when the neutrality output conditions are met.

[0596] The neutrality evaluation means 53411 obtains, for example, one or more neutrality output conditions from the condition management unit 512. Next, the neutrality evaluation means 53411 determines whether any of the obtained neutrality output conditions are met.

[0597] The settlement proposal means 53412 performs a settlement proposal generation process. That is, the settlement proposal means 53412, for example, provides the generation AI with a settlement proposal generation AI agent configuration prompt, and configures the settlement proposal generation AI agent based on the prompt. Then, the settlement proposal means 53412 performs a settlement proposal generation process. The settlement proposal generation process is the process of generating and outputting a settlement proposal. The settlement proposal means 53412 provides the settlement proposal generation AI agent with one or more types of information, for example, negotiation information entered by the petitioner, negotiation information entered by the respondent, negotiation information entered by a third party, or generated information output by any of the AI ​​agents, and obtains a response from the settlement proposal generation AI agent that includes one or more settlement proposals. The settlement proposal means 53412 outputs the one or more settlement proposals to one or more users, including the petitioner, respondent, or third party.

[0598] The settlement proposal means 53412 searches a set of static information and / or a set of dynamic information using, for example, claim information, or negotiation information received during negotiations, or dialogue information indicating the content of conversations between the participants or other AI agents and the settlement proposal generation AI agent during negotiations, or generated information output by other AI agents during negotiations, and obtains one or more pieces of information. Next, the agent means 5341 passes, for example, the one or more pieces of information to the settlement proposal generation AI agent.

[0599] The settlement proposal means 53412 is preferable to output a settlement proposal when the conditions for creating a settlement proposal are met.

[0600] The decision unit 537 performs agent decision processing. The decision unit 537 also performs settlement proposal decision processing. Furthermore, the decision unit 537 performs, for example, the processing performed by the decision unit 137.

[0601] The agent determination process is the process of determining the type of AI agent to use. For example, the agent determination process is the process of obtaining one or more agent type identifiers.

[0602] The agent determination process is, for example, the process of obtaining one or more agent type identifiers contained in the information or instructions received from the petitioner's terminal 2, the respondent's terminal 3, or the third-party terminal 6. The agent determination process is, for example, the process of determining whether one or more agent linkage conditions are met and obtaining the agent type identifiers that are paired with the agent linkage conditions that are determined to be met.

[0603] The settlement proposal judgment process is the process of deciding whether or not to present a settlement proposal. The settlement proposal judgment process is the process of determining whether the application status information, or information based on the application status information, that is paired with the application identifier of the application that is the subject of the settlement proposal, satisfies the conditions for creating a settlement proposal. For example, the settlement proposal judgment process is the process of determining whether the application status information, or information based on the application status information, that is paired with the application identifier of the application that is the subject of the settlement proposal, satisfies one or more of the conditions for creating a settlement proposal.

[0604] The transmitting unit 54 transmits various types of information. These types of information include, for example, neutrality assessment results, settlement proposals, notification information, screen information, proposed content, negotiation information, dialogue information, necessary information, and agreement documents.

[0605] The third-party storage unit 61, which constitutes the third-party terminal 6, stores various types of information. These types of information include, for example, a claim identifier and a third-party identifier. The claim identifier in the third-party storage unit 61 is an identifier for claim information that identifies the dispute supported by the third party. The third-party identifier is information that identifies the third party.

[0606] The third-party reception unit 62 receives various types of information and instructions. These types of information and instructions include, for example, negotiation information, third-party input, proxy instructions, and support instructions. The third-party reception unit 62 receives, for example, phone calls from the AI ​​agent or the negotiation support device 5.

[0607] The third-party processing unit 63 performs various processes. These processes include, for example, converting received information and instructions into information and instructions for transmission. Other processes include, for example, converting received information into information for output.

[0608] The third-party transmission unit 64 transmits various information and instructions to the negotiation support device 5 or other devices (not shown). These various information and instructions include, for example, negotiation information, third-party input, proxy instructions, and support instructions.

[0609] The third-party receiving unit 65 receives various types of information. These types of information include, for example, negotiation information, claim status information, or various screen information.

[0610] The third-party output unit 66 outputs various types of information. These types of information include, for example, negotiation information, application status information, various screens, or voice information from the AI ​​agent.

[0611] Here, "output" is a concept that includes display on a screen, projection using a projector, printing with a printer, sound output, transmission to an external device, storage on a recording medium, and transfer of processing results to other processing devices or other programs.

[0612] The storage unit 51, the claim management unit 511, the condition management unit 512, the prompt management unit 514, and the third-party storage unit 61 are preferably made of non-volatile recording media, but can also be made of volatile recording media.

[0613] The process by which information is stored in the storage unit 51, etc. is not relevant. For example, information may be stored in the storage unit 51, etc. via a recording medium, information transmitted via a communication line, etc. may be stored in the storage unit 51, etc., or information input via an input device may be stored in the storage unit 51, etc.

[0614] The receiving unit 52, the instruction receiving unit 524, and the third-party receiving unit 65 are usually implemented by wireless or wired communication means, but may also be implemented by means of receiving broadcasts.

[0615] The processing unit 53, negotiation support unit 534, agent means 5341, neutrality evaluation means 53411, settlement proposal means 53412, judgment unit 537, and third-party processing unit 63 can typically be implemented using a processor, memory, etc. The processing procedures of the processing unit 53, etc., are typically implemented in software, and this software is recorded on a recording medium such as ROM. However, it may also be implemented in hardware (dedicated circuitry). The processor can be a CPU, MPU, GPU, etc., and the type is not limited.

[0616] The transmitting unit 54 and the third-party transmitting unit 64 are usually implemented by wireless or wired communication means, but may also be implemented by broadcasting means.

[0617] The third-party reception unit 62 can be implemented using device drivers for input means such as touch panels and keyboards, or control software for menu screens, etc.

[0618] The third-party output unit 66 may or may not be considered to include output devices such as displays and speakers. The third-party output unit 66 can be implemented using driver software for an output device, or by using driver software for an output device and an output device.

[0619] Next, an example of the operation of the negotiation support device 5 will be explained using the flowcharts in Figures 36 to 38. In the flowcharts in Figures 36 to 38, the explanation of the same steps as in the flowcharts in Figures 4 to 6 will be omitted.

[0620] In addition, in the flowcharts shown in Figures 36 to 38, each of the one or more AI agents used may be configured using the AI ​​agent configuration prompt before the start of operation of the negotiation support device 5. Furthermore, when configuring an AI agent, it is preferable to provide the AI ​​agent with one or more types of static information and one or more types of dynamic information.

[0621] (Step S3601) The petition acquisition unit 121 determines whether or not it has received a case registration from the petitioner terminal 2. If a case registration is received, it proceeds to step S411; otherwise, it proceeds to step S3609. A case registration is an instruction to register the case that is the subject of the petition. A case registration here includes, for example, petition information and one or more agent type identifiers.

[0622] The agent type identifier here is information that identifies the type of AI agent that the claimant who submitted the case registration wishes to use.

[0623] (Step S3602) The determination unit 537 performs agent determination processing. An example of agent determination processing will be explained using the flowchart in Figure 39.

[0624] (Step S3603) If the determination unit 537 obtains the agent type identifier "Petitioner's Representative" in step S3602, the process proceeds to step S3604; otherwise, the process proceeds to step S3605.

[0625] (Step S3604) Agent means 5341 performs the petitioner's proxy processing. Return to step S3601. An example of petitioner's proxy processing will be explained using the flowchart in Figure 40.

[0626] (Step S3605) If the determination unit 537 obtains the agent type identifier "petitioner support" in step S3602, it proceeds to step S3606; otherwise, it returns to step S3601.

[0627] (Step S3606) Agent means 5341 performs petitioner support processing. An example of petitioner support processing will be explained using the flowchart in Figure 41.

[0628] (Step S3607) The processing unit 53 determines whether the result of the judgment in step S3606 "meets the notification conditions". If it "meets the notification conditions", proceed to step S414; if it "does not meet the notification conditions", proceed to step S3608. If it "does not meet the notification conditions", the applicant withdraws the application.

[0629] (Step S3608) The processing unit 53 obtains a claim identifier that identifies the claim in question. The processing unit 53 deletes the claim information that is paired with the claim identifier from the claim management unit 511. The process returns to step S3601. As a result of deleting such claim information, the received case registration is effectively nullified. In other words, with the assistance of the claimant support AI agent, the claimant has decided not to file a claim.

[0630] (Step S3609) The confirmation receiving unit 122 determines whether or not it has received confirmation information, etc. from the respondent terminal 3. If confirmation information, etc. has been received, the unit proceeds to step S421; otherwise, it proceeds to step S3615. Confirmation information, etc. refers to confirmation information or non-existent information. The received confirmation information, etc. is associated with the claim identifier. Here, the confirmation information, etc. has, for example, one or more agent type identifiers.

[0631] The agent type identifier here is information that identifies the type of AI agent that the respondent who sent the confirmation information, etc., wishes to use.

[0632] (Step S3610) The determination unit 537 performs agent determination processing. An example of agent determination processing will be explained using the flowchart in Figure 39.

[0633] (Step S3611) In step S3610, if the determination unit 537 obtains the agent type identifier "Respondent's Representative", it proceeds to step S3612; otherwise, it proceeds to step S3613.

[0634] (Step S3612) Agent means 5341 performs the respondent's proxy processing. An example of the respondent's proxy processing will be explained using the flowchart in Figure 42.

[0635] (Step S3613) If the determination unit 537 obtains the agent type identifier "respondent support" in step S3610, it proceeds to step S3614, and if it does not obtain "respondent support", it returns to step S3601.

[0636] (Step S3614) Agent means 5341 performs respondent support processing. An example of respondent support processing will be explained using the flowchart in Figure 43.

[0637] (Step S3615) The negotiation receiving unit 123 determines whether it has received negotiation information, etc. from the petitioner terminal 2, the respondent terminal 3, or the third-party terminal 6. If negotiation information, etc. has been received, the unit proceeds to step S3616; otherwise, it proceeds to step S3617. The received negotiation information, etc. is associated with the petition identifier. Furthermore, the negotiation information, etc. may consist only of negotiation information, or it may include negotiation information and one or more AI agent type identifiers.

[0638] (Step S3616) The negotiation support unit 534 performs negotiation information reception processing. Return to step S3601. An example of negotiation information reception processing will be explained using the flowchart in Figure 44.

[0639] (Step S3617) The instruction receiving unit 524 determines whether or not it has received an instruction from the applicant's terminal 2. If an instruction from the applicant's terminal has been received, the unit proceeds to step S3618; otherwise, it proceeds to step S3619.

[0640] (Step S3618) Agent means 5341 performs the petitioner's proxy processing. Return to step S3601. An example of petitioner's proxy processing will be explained using the flowchart in Figure 40.

[0641] (Step S3619) The instruction receiving unit 524 determines whether or not it has received an instruction to support the petitioner from the petitioner terminal 2. If an instruction to support the petitioner has been received, the unit proceeds to step S3620; otherwise, it proceeds to step S3621.

[0642] (Step S3620) Agent means 5341 performs petitioner support processing. Return to step S3601. An example of petitioner support processing will be explained using the flowchart in Figure 41.

[0643] (Step S3621) The instruction receiving unit 524 determines whether or not it has received an instruction to act on behalf of the respondent from the respondent terminal 3. If an instruction to act on behalf of the respondent has been received, the unit proceeds to step S3622; otherwise, the unit proceeds to step S3623.

[0644] (Step S3622) Agent means 5341 performs the respondent's proxy processing. Return to step S3601. An example of the respondent's proxy processing will be explained using the flowchart in Figure 42.

[0645] (Step S3623) The instruction receiving unit 524 determines whether or not it has received an instruction to act on behalf of the respondent from the respondent terminal 3. If an instruction to act on behalf of the respondent has been received, the unit proceeds to step S3624; otherwise, the unit proceeds to step S3625.

[0646] (Step S3624) Agent means 5341 performs respondent support processing. Return to step S3601. An example of respondent support processing will be explained using the flowchart in Figure 43.

[0647] (Step S3625) The instruction receiving unit 524 determines whether it has received a third-party proxy instruction from the petitioner terminal 2, the respondent terminal 3, or the third-party terminal 6. If a third-party proxy instruction is received, the unit proceeds to step S3626; otherwise, it proceeds to step S3627.

[0648] (Step S3626) Agent means 5341 performs third-party proxy processing. Return to step S3601. An example of third-party proxy processing will be explained using the flowchart in Figure 45.

[0649] (Step S3627) The instruction receiving unit 524 determines whether or not it has received a third-party support instruction from the third-party terminal 6. If a third-party support instruction is received, the unit proceeds to step S3628; otherwise, it proceeds to step S440.

[0650] (Step S3628) Agent means 5341 performs third-party support processing. Return to step S3601. An example of third-party support processing will be explained using the flowchart in Figure 46.

[0651] In the flowcharts shown in Figures 36 to 38, processing is terminated by power-off or processing termination interrupts.

[0652] Next, an example of the agent decision process in steps S3602, S3610, and S4401 will be explained using the flowchart in Figure 39.

[0653] (Step S3901) The determination unit 537 determines whether an agent type identifier exists in the information or instructions received from the petitioner terminal 2, the respondent terminal 3, or the third-party terminal 6. If an agent type identifier exists, the unit proceeds to step S3902; otherwise, it proceeds to step S3903.

[0654] (Step S3902) The decision unit 537 obtains one or more agent type identifiers from the received information or instructions. The decision unit 537 obtains the claim identifier of the claim in question. The decision unit 537 stores the obtained one or more agent type identifiers in association with the claim identifier. The decision unit returns to the higher-level processing.

[0655] (Step S3903) The determination unit 537 assigns 1 to counter i.

[0656] (Step S3904) The determination unit 537 determines whether the i-th agent linkage condition exists in the condition management unit 512. If the i-th agent linkage condition exists, the process proceeds to step S3905; otherwise, the process proceeds to step S3910.

[0657] (Step S3905) The determination unit 537 obtains the i-th agent cooperation condition from the condition management unit 512.

[0658] (Step S3906) The decision unit 537 acquires one or more pieces of information necessary to determine the i-th agent collaboration condition acquired in step S3905. Note that a module for acquiring the information necessary for the determination may be associated with the agent collaboration condition, or information identifying the information necessary for the determination may be associated with the agent collaboration condition, or the decision unit 537 may analyze the agent collaboration condition and acquire information identifying the information necessary for determining the condition. The method for acquiring one or more pieces of information necessary to determine the i-th agent collaboration condition is not limited.

[0659] (Step S3907) The determination unit 537 determines whether the one or more pieces of information obtained in step S3906 satisfy the i-th agent cooperation condition. If the agent cooperation condition is satisfied, the unit proceeds to step S3908; otherwise, the unit proceeds to step S3909.

[0660] (Step S3908) The decision unit 537 obtains the agent type identifier that corresponds to the i-th agent linkage condition from the condition management unit 512. The decision unit 537 obtains the application identifier of the application in question. The decision unit 537 stores the obtained agent type identifier in association with the said application identifier.

[0661] (Step S3909) The determination unit 537 increments counter i by 1. Return to step S3904.

[0662] In the flowchart shown in Figure 39, if all agent collaboration conditions are not met, the agent type identifier will not be stored.

[0663] Furthermore, in the flowchart of Figure 39, if it is determined that one agent linkage condition is met, the agent type identifier corresponding to that agent linkage condition may be obtained from the condition management unit 512, the agent type identifier may be stored in association with the claim identifier, and the process may be returned to the higher level.

[0664] Next, we will explain examples of the petitioner's proxy processing in steps S3604, S3618, and S4403 using the flowchart in Figure 40.

[0665] (Step S4001) The agent means 5341 obtains a claim identifier that identifies the claim in question.

[0666] (Step S4002) The agent means 5341 determines whether or not there is an agent type identifier "Petitioner AI Agent" that corresponds to the petition identifier obtained in step S4001, based on the agent management information paired with the petition identifier in the petition management unit 511. If a petitioner AI Agent exists, the process proceeds to step S4004; otherwise, the process proceeds to step S4003.

[0667] (Step S4003) Agent means 5341 performs the petitioner AI agent configuration process.

[0668] An example of the petitioner AI agent configuration process will be explained using the flowchart in Figure 47. In step S4003, the agent means 5341 passes an agent type identifier that identifies the petitioner AI agent to the module that performs the AI ​​agent configuration process.

[0669] (Step S4004) The agent means 5341 obtains one or more pieces of information from the claim management unit 511 that are paired with the claim identifier obtained in step S4001.

[0670] Furthermore, it is preferable that the agent means 5341 does not acquire information that has already been provided to the petitioner AI agent. The one or more pieces of information may include, for example, petition information, negotiation information, dialogue information, one or more types of static information, and one or more types of dynamic information.

[0671] (Step S4005) Agent means 5341 passes one or more pieces of information obtained in step S4004 to the claimant AI agent.

[0672] Furthermore, the agent means 5341 stores identifiers for information indicating that it has passed each of the one or more pieces of information, associated with the claimant AI agent (for example, associated with the agent type identifier "claimant AI agent" which is paired with the claim identifier).

[0673] (Step S4006) Agent means 5341 determines whether or not it has received a response from the claimant AI agent. If a response has been received, it proceeds to step S4007; otherwise, it returns to step S4006.

[0674] Furthermore, responses from the AI ​​agent can be considered as notifications of execution results from the AI ​​agent, outputs from the AI ​​agent, processing results from the AI ​​agent, generated results from the AI ​​agent, etc. The response is interpreted broadly.

[0675] (Step S4007) Agent means 5341 obtains generated information from the response obtained in step S4006 and stores it in association with the claim identifier and the agent type identifier "claimant AI agent". The generated information may be the same information as the response, or it may be part of the information of the response. Agent means 5341 may also store the time obtained from a clock (not shown) in association with the claim identifier.

[0676] (Step S4008) Agent means 5341 obtains contact information of one or more interlocutors to whom generated information is conveyed from the claim management unit 511. Interlocutors here typically include the respondent. Interlocutors may also include the claimant or a third party. Interlocutors include, for example, the respondent's agent and the third-party agent. The contact information here may include information for passing information to the AI ​​agent (for example, an API for accessing the AI ​​agent).

[0677] For example, the agent means 5341 may obtain agent management information paired with the claim identifier and include AI agents other than the claimant AI agent among the one or more agent type identifiers indicated by the agent management information as conversation partners.

[0678] (Step S4009) The agent means 5341 uses the contact information obtained in step S4008 to send generated information to one or more conversation partners.

[0679] If the conversation partner is an AI agent, the agent means 5341 performs the process of passing the generated information to the AI ​​agent. In other words, the process of sending generated information to an AI agent is the process of passing the generated information to the AI ​​agent.

[0680] (Step S4010) Agent means 5341 decides whether to terminate the dialogue or not. If the dialogue is to be terminated, it returns to the higher-level process; otherwise, it proceeds to step S4011.

[0681] In addition, a dialogue will be terminated if, for example, the conditions for terminating the dialogue are met. These conditions include, for example, that the dialogue information from the other party indicates the "end" of the dialogue, that the response from the claimant AI agent indicates the "end" of the dialogue, that a command to terminate the dialogue has been received, or that no dialogue information has been received from the other party for a specified period of time, but these conditions are not strictly enforced.

[0682] (Step S4011) The agent means 5341 determines whether or not it has received dialogue information, which is information from one or more dialogue partners. If dialogue information has been received, it proceeds to step S4012; otherwise, it returns to step S4011.

[0683] (Step S4012) The agent means 5341 stores the one or more dialogue information received in step S4011 in the claim management unit 511, associating it with the claim identifier and the identifier of the person who sent the dialogue information. Here, it is preferable for the agent means 5341 to obtain the time from a clock (not shown) and store the time as associated with the dialogue information.

[0684] (Step S4013) Agent means 5341 passes one or more conversational pieces of information to the claimant AI agent. Return to step S4006.

[0685] In addition, in the flowchart of Figure 40, if no dialogue information is received from the dialogue partner for a predetermined period of time or longer in step S4011, the dialogue process may be terminated and the process may return to a higher level.

[0686] Furthermore, in the flowchart of Figure 40, there may be two or more parties with whom the petitioner's AI agent interacts. Two or more parties with whom interaction is possible include, for example, the respondent, one or more third parties, the respondent's AI agent, and one or more third-party AI agents, with two or more of these being included.

[0687] Next, examples of petitioner support processing in steps S3606, S3620, and S4405 will be explained using the flowchart in Figure 41.

[0688] (Step S4101) The agent means 5341 obtains a claim identifier that identifies the claim in question.

[0689] (Step S4102) The agent means 5341 determines whether or not an agent type identifier "Petitioner Support AI Agent" corresponding to the petition identifier exists, based on the agent management information paired with the petition identifier in the petition management unit 511. If a petitioner support AI agent exists, the process proceeds to step S4104; otherwise, the process proceeds to step S4103.

[0690] (Step S4103) Agent means 5341 performs petitioner support AI agent configuration processing.

[0691] An example of the petitioner support AI agent configuration process will be explained using the flowchart in Figure 47. In step S4103, the agent means 5341 passes an agent type identifier that identifies the petitioner support AI agent to the module that performs the AI ​​agent configuration process.

[0692] (Step S4104) The agent means 5341 obtains one or more pieces of information from the claim management unit 511 that are paired with the claim identifier obtained in step S4101.

[0693] Furthermore, it is preferable that the agent means 5341 does not acquire information that has already been provided to the petitioner support AI agent. The one or more pieces of information include, for example, petition information, negotiation information, dialogue information, one or more types of static information, and one or more types of dynamic information.

[0694] (Step S4105) Agent means 5341 passes one or more pieces of information obtained in step S4104 to the petitioner support AI agent.

[0695] (Step S4106) Agent means 5341 determines whether or not it has obtained a response from the petitioner support AI agent. If a response has been obtained, it proceeds to step S4107; otherwise, it returns to step S4106.

[0696] (Step S4107) Agent means 5341 obtains generated information from the response obtained in step S4106 and stores it in association with the claim identifier and the agent type identifier. Agent means 5341 may also store the time obtained from a clock (not shown) in association with the claim identifier.

[0697] (Step S4108) Agent means 5341 obtains contact information of one or more parties to communicate the generated information from the Claims Management Unit 511. The parties to communicate usually include the claimant. The parties to communicate usually do not include the respondent. The parties to communicate may include third parties. The parties to communicate include, for example, the claimant's agent and the third-party agent.

[0698] The agent means 5341 may, for example, obtain agent management information paired with a claim identifier, and preferably include AI agents other than the claimant support AI agent, the respondent AI agent, and the respondent support AI agent among the one or more agent type identifiers indicated by the agent management information as conversation partners.

[0699] (Step S4109) The agent means 5341 uses the contact information obtained in step S4108 to send generated information to one or more conversation partners. If the conversation partner is an AI agent, the agent means 5341 performs the process of passing the generated information to the AI ​​agent.

[0700] (Step S4110) The agent means 5341 decides whether or not to terminate the dialogue. If it decides to terminate the dialogue, it goes to step S4114; otherwise, it goes to step S4011. The dialogue is terminated, for example, when the above-mentioned dialogue termination conditions are met.

[0701] (Step S4111) Agent means 5341 determines whether or not it has received dialogue information from the claimant. If dialogue information has been received, it proceeds to step S4112; otherwise, it returns to step S4111.

[0702] (Step S4112) The agent means 5341 associates the dialogue information received in step S4111 with the claim identifier and the claimant's identifier and stores it in the claim management unit 511. Here, it is preferable for the agent means 5341 to obtain the time from a clock (not shown) and store the time associated with the dialogue information.

[0703] (Step S4113) Agent means 5341 passes the received dialogue information to the claimant support AI agent. Return to step S4106.

[0704] (Step S4114) The agent means 5341 refers to the application status information paired with the application identifier and determines whether or not it has sent notification information to the respondent. If it has sent notification information, it proceeds to step S4115; otherwise, it proceeds to step S4116.

[0705] (Step S4115) Agent means 5341 assigns "not met (e.g., "0")" to the variable "notification condition result". It returns to the higher-level process.

[0706] (Step S4116) The agent means 5341 decides whether or not to notify the respondent based on one or more conversational pieces of information sent from the petitioner. If it decides not to notify the respondent, it proceeds to step S4115; if it decides not to notify, it proceeds to step S4117. If it decides not to notify, for example, one or more conversational pieces of information sent from the petitioner are classified into a class that means "I will not file a petition" or "I will withdraw my petition." The method of classifying the information is not limited. Possible methods of classifying the information include machine learning methods, methods using generative AI, and methods using natural reduction processing, and these are publicly known technologies.

[0707] (Step S4117) Agent means 5341 assigns "satisfied" (e.g., "1") to the variable "notification condition result". It returns to the higher-level process.

[0708] Next, we will explain examples of the respondent's proxy processing in steps S3612, S3622, and S4407 using the flowchart in Figure 42.

[0709] (Step S4201) The agent means 5341 obtains a claim identifier that identifies the claim in question.

[0710] (Step S4202) The agent means 5341 determines whether or not an agent type identifier "Respondent AI Agent" exists that corresponds to the claim identifier obtained in step S4201, based on the agent management information paired with the claim identifier in the claim management unit 511. If a Respondent AI Agent exists, the process proceeds to step S4204; otherwise, the process proceeds to step S4203.

[0711] (Step S4203) Agent means 5341 performs respondent AI agent configuration processing.

[0712] An example of the respondent AI agent configuration process will be explained using the flowchart in Figure 47. In step S4203, the agent means 5341 passes an agent type identifier that identifies the respondent AI agent to the module that performs the AI ​​agent configuration process.

[0713] (Step S4204) The agent means 5341 obtains one or more pieces of information from the claim management unit 511 that are paired with the claim identifier obtained in step S4201.

[0714] Furthermore, it is preferable that the agent means 5341 does not acquire information that has already been provided to the respondent AI agent. The one or more pieces of information include, for example, petition information, negotiation information, dialogue information, one or more types of static information, and one or more types of dynamic information.

[0715] (Step S4205) Agent means 5341 passes one or more pieces of information obtained in step S4204 to the respondent AI agent.

[0716] (Step S4206) Agent means 5341 determines whether or not it has obtained a response from the respondent AI agent. If a response has been obtained, it proceeds to step S4207; otherwise, it returns to step S4206.

[0717] (Step S4207) Agent means 5341 obtains generated information from the response obtained in step S4206 and stores it in association with the claim identifier and the agent type identifier. Agent means 5341 may also store the time obtained from a clock (not shown) in association with the claim identifier.

[0718] (Step S4208) Agent means 5341 obtains contact information of one or more parties to communicate the generated information from the Claims Management Unit 511. The parties to communicate typically include the claimant. The parties to communicate may also include the respondent or a third party. The parties to communicate include, for example, the claimant's agent and the third-party agent.

[0719] For example, the agent means 5341 may obtain agent management information paired with a claim identifier, and preferably include AI agents other than the respondent AI agent among the one or more agent type identifiers indicated by the agent management information as conversation partners.

[0720] (Step S4209) The agent means 5341 uses the contact information obtained in step S4208 to send generated information to one or more conversation partners.

[0721] If the conversation partner is an AI agent, the agent means 5341 performs the process of passing the generated information to the AI ​​agent.

[0722] (Step S4210) The agent means 5341 decides whether or not to terminate the dialogue. If it decides to terminate the dialogue, it returns to the higher-level process; otherwise, it proceeds to step S4211. The dialogue is terminated, for example, when the above-mentioned dialogue termination conditions are met.

[0723] (Step S4211) The agent means 5341 determines whether or not it has received dialogue information, which is information from one or more dialogue partners. If dialogue information has been received, the process proceeds to step S4212; otherwise, it returns to step S4211.

[0724] (Step S4212) The agent means 5341 stores the one or more dialogue information received in step S4211 in the claim management unit 511, associating the claim identifier with the identifier of the dialogue partner. Here, it is preferable for the agent means 5341 to obtain the time from a clock (not shown) and store the time as associated with the dialogue information.

[0725] (Step S4213) Agent means 5341 passes one or more conversational pieces of information to the respondent AI agent. Return to step S4206.

[0726] In addition, in the flowchart of Figure 42, if no dialogue information is received from the dialogue partner for a predetermined period of time or longer in step S4211, the dialogue process may be terminated and the process may return to a higher level.

[0727] Furthermore, in the flowchart of Figure 42, there may be two or more parties with whom the respondent AI agent interacts. Two or more parties with whom interaction is possible include, for example, the petitioner, one or more third parties, the petitioner's AI agent, and one or more third-party AI agents, with two or more of these being the parties involved.

[0728] Next, we will explain examples of respondent support processing in steps S3614, S3624, and S4409 using the flowchart in Figure 43.

[0729] (Step S4301) The agent means 5341 obtains a claim identifier that identifies the claim in question.

[0730] (Step S4302) The agent means 5341 determines whether or not an agent type identifier "Respondent Support AI Agent" exists that corresponds to the claim identifier obtained in step S4301, based on the agent management information paired with the claim identifier in the claim management unit 511. If a Respondent Support AI Agent exists, the process proceeds to step S4304; otherwise, the process proceeds to step S4303.

[0731] (Step S4303) Agent means 5341 performs respondent support AI agent configuration processing.

[0732] An example of the respondent support AI agent configuration process will be explained using the flowchart in Figure 47. In step S4303, the agent means 5341 passes an agent type identifier that identifies the respondent support AI agent to the module that performs the AI ​​agent configuration process.

[0733] (Step S4304) The agent means 5341 obtains one or more pieces of information from the claim management unit 511 that are paired with the claim identifier obtained in step S4301.

[0734] Furthermore, it is preferable that the agent means 5341 does not acquire information that has already been provided to the respondent support AI agent. The one or more pieces of information include, for example, petition information, negotiation information, dialogue information, one or more types of static information, and one or more types of dynamic information.

[0735] (Step S4305) Agent means 5341 passes one or more pieces of information obtained in step S4304 to the respondent support AI agent.

[0736] (Step S4306) Agent means 5341 determines whether or not it has obtained a response from the respondent support AI agent. If a response has been obtained, it proceeds to step S4307; otherwise, it returns to step S4306.

[0737] (Step S4307) Agent means 5341 obtains generated information from the response obtained in step S4306 and stores it in association with the claim identifier and the agent type identifier. Agent means 5341 may also store the time obtained from a clock (not shown) in association with the claim identifier.

[0738] (Step S4308) Agent means 5341 obtains contact information of one or more parties to communicate the generated information from the Claims Management Unit 511. The parties to communicate usually include the respondent. The parties to communicate usually do not include the claimant. The parties to communicate may include third parties. The parties to communicate include, for example, the respondent's agent and the third-party agent.

[0739] For example, the agent means 5341 may acquire agent management information paired with a claim identifier, and preferably include respondent support AI agents, claimant AI agents, and AI agents excluding the claimant support AI agent as conversation partners among the one or more agent type identifiers indicated by the agent management information.

[0740] (Step S4309) The agent means 5341 uses the contact information obtained in step S4308 to send generated information to one or more conversation partners. If the conversation partner is an AI agent, the agent means 5341 performs the process of passing the generated information to the AI ​​agent.

[0741] (Step S4310) The agent means 5341 decides whether or not to terminate the dialogue. If it decides to terminate the dialogue, it returns to the higher-level process; otherwise, it proceeds to step S4011. The dialogue is terminated, for example, when the above-mentioned dialogue termination conditions are met.

[0742] (Step S4311) Agent means 5341 determines whether or not it has received dialogue information from the respondent. If dialogue information has been received, it proceeds to step S4312; otherwise, it returns to step S4311.

[0743] (Step S4312) The agent means 5341 associates the dialogue information received in step S4311 with the claim identifier and the respondent identifier and stores it in the claim management unit 511. Here, it is preferable for the agent means 5341 to obtain the time from a clock (not shown) and store the time in association with the dialogue information.

[0744] (Step S4313) Agent means 5341 passes the received dialogue information to the respondent support AI agent. Return to step S4306.

[0745] Next, an example of the negotiation information reception process in step S3616 will be explained using the flowchart in Figure 44. In the flowchart in Figure 44, the explanation of steps identical to those in the flowchart in Figure 5 will be omitted.

[0746] (Step S4401) The determination unit 537 performs agent determination processing. An example of agent determination processing will be explained using the flowchart in Figure 39.

[0747] (Step S4402) If the determination unit 537 has obtained the agent type identifier "petitioner's representative" in step S4401, the process proceeds to step S4403; otherwise, the process proceeds to step S4404.

[0748] (Step S4403) Agent means 5341 performs the petitioner's proxy processing. An example of petitioner's proxy processing will be explained using the flowchart in Figure 40.

[0749] (Step S4404) If the determination unit 537 has obtained the agent type identifier "petitioner support" in step S4401, proceed to step S4405; otherwise, proceed to step S4406.

[0750] (Step S4405) Agent means 5341 performs petitioner support processing. An example of petitioner support processing will be explained using the flowchart in Figure 41.

[0751] (Step S4406) If the determination unit 537 has obtained the agent type identifier "Respondent's Representative" in step S4401, it proceeds to step S4407; if it has not obtained "Petitioner Support", it proceeds to step S4408.

[0752] (Step S4407) Agent means 5341 performs the respondent's proxy processing. An example of the respondent's proxy processing will be explained using the flowchart in Figure 42.

[0753] (Step S4408) If the determination unit 537 has obtained the agent type identifier "respondent support" in step S4401, it proceeds to step S4409; if it has not obtained "petitioner support", it proceeds to step S4410.

[0754] (Step S4409) Agent means 5341 performs respondent support processing. An example of respondent support processing will be explained using the flowchart in Figure 43.

[0755] (Step S4410) If the determination unit 537 has obtained the agent type identifier "Third-party proxy" in step S4401, it proceeds to step S4411; otherwise, it proceeds to step S4412.

[0756] (Step S4411) The agent means 5341 performs third-party proxy processing. An example of third-party proxy processing will be explained using the flowchart in Figure 45.

[0757] (Step S4412) If the determination unit 537 has obtained the agent type identifier "Third-party support" in step S4401, it proceeds to step S4413; otherwise, it proceeds to step S4419.

[0758] (Step S4413) The agent means 5341 performs third-party support processing. An example of third-party support processing will be explained using the flowchart in Figure 46.

[0759] (Step S4414) The decision unit 537 performs a settlement proposal decision process. An example of the settlement proposal decision process will be explained using the flowchart in Figure 48.

[0760] (Step S4415) If the result of the decision in step S4414 is to "create a settlement proposal (e.g., "1")", proceed to step S4416; if it is to "not create a settlement proposal (e.g., "0")", return to the higher-level process.

[0761] (Step S4416) The settlement proposal means 53412 performs a settlement proposal generation process and obtains one or more settlement proposals. An example of the settlement proposal generation process will be explained using the flowchart in Figure 49.

[0762] (Step S4417) The settlement proposal means 53412 obtains one or more contact details to identify recipients to whom the settlement proposal will be sent. Recipients may be, for example, the petitioner and the respondent. Recipients may also include third parties. Recipients may include, for example, one or more AI agents.

[0763] The settlement proposal means 53412 preferably obtains agent management information paired with a claim identifier and includes one or more AI agents identified by one or more agent type identifiers indicated by the agent management information as the notification recipient.

[0764] (Step S4418) The settlement proposal means 53412 sends one or more settlement proposals obtained in step S4416 to one or more recipients using one or more contact information. It then returns to the higher-level process.

[0765] (Step S4419) The neutrality evaluation means 53411 determines whether the neutrality evaluation conditions are met. If the neutrality evaluation conditions are met, the process proceeds to step S4420; otherwise, it returns to the higher-level process.

[0766] (Step S4420) The neutrality evaluation means 53411 performs a neutrality evaluation process and obtains the neutrality evaluation result. An example of the neutrality evaluation process will be explained using the flowchart in Figure 50.

[0767] (Step S4421) The determination unit 537 determines whether the neutrality evaluation result satisfies the neutrality output condition. If the neutrality output condition is met, the process proceeds to step S4422; otherwise, it returns to the higher-level process.

[0768] (Step S4422) The neutrality evaluation means 53411 transmits the neutrality evaluation result. It returns to the higher-level processing.

[0769] Furthermore, the transmission of neutrality assessment results may include, for example, transmission to a third party. This could involve providing the results to a third-party AI agent or a third-party support AI agent.

[0770] Furthermore, in the flowchart of Figure 44, if the result in step S4415 is "N", the process may proceed from step S4418 to step S4419. In other words, the timing of determining whether the neutrality assessment conditions are met and the timing of performing the neutrality assessment process are not specified. Also, in the flowchart of Figure 44, the timing of judging the settlement proposal and the timing of generating the settlement proposal are not specified.

[0771] Next, an example of third-party proxy processing in step S3626 will be explained using the flowchart in Figure 45.

[0772] (Step S4501) The agent means 5341 obtains a claim identifier that identifies the claim in question.

[0773] (Step S4502) The agent means 5341 determines whether or not there is an agent type identifier "Third-Party AI Agent" that corresponds to the claim identifier obtained in step S4501, based on the agent management information paired with the claim identifier in the claim management unit 511. If a third-party AI agent exists, the process proceeds to step S4504; otherwise, the process proceeds to step S4503.

[0774] (Step S4503) Agent means 5341 performs third-party AI agent configuration processing.

[0775] An example of third-party AI agent configuration processing will be explained using the flowchart in Figure 47. In step S4503, the agent means 5341 passes an agent type identifier that identifies the third-party AI agent to the module that performs AI agent configuration processing.

[0776] (Step S4504) The agent means 5341 obtains one or more pieces of information from the claim management unit 511 that are paired with the claim identifier obtained in step S4501.

[0777] Furthermore, it is preferable that the agent means 5341 does not acquire information that has already been provided to a third-party AI agent. The one or more pieces of information include, for example, claim information, negotiation information, dialogue information, one or more types of static information, and one or more types of dynamic information.

[0778] (Step S4505) The agent means 5341 passes one or more pieces of information obtained in step S4504 to a third-party AI agent.

[0779] (Step S4506) Agent means 5341 determines whether or not it has obtained a response from the third-party AI agent. If a response has been obtained, it proceeds to step S4507; otherwise, it returns to step S4506.

[0780] (Step S4507) Agent means 5341 obtains generated information from the response obtained in step S4506 and stores it in association with the claim identifier and the agent type identifier. Agent means 5341 may also store the time obtained from a clock (not shown) in association with the claim identifier.

[0781] (Step S4508) The neutrality evaluation means 53411 uses the generated information obtained in step S4507 to perform a neutrality evaluation process and obtain a neutrality evaluation result. An example of the neutrality evaluation process will be explained using the flowchart in Figure 50.

[0782] (Step S4509) The determination unit 537 determines whether the neutrality evaluation result in step S4508 satisfies the neutrality output condition. If the neutrality output condition is met, the process proceeds to step S4510; otherwise, the process proceeds to step S4511.

[0783] (Step S4510) The neutrality assessment means 53411 passes the neutrality assessment result or information based on the neutrality assessment result (e.g., "a prompt to provide a neutral response") to the third-party AI agent. Return to step S4506.

[0784] (Step S4511) Agent means 5341 obtains contact information of one or more parties to communicate the generated information from the Claims Management Unit 511. The parties to communicate typically include the claimant. The parties to communicate may also include the respondent or a third party. The parties to communicate include, for example, the claimant's agent and the third-party agent.

[0785] For example, the agent means 5341 may obtain agent management information paired with a claim identifier, and preferably include AI agents, excluding third-party AI agents, among the one or more agent type identifiers indicated by the agent management information as conversation partners.

[0786] (Step S4512) The agent means 5341 uses the contact information obtained in step S4508 to send generated information to one or more conversation partners.

[0787] If the conversation partner is an AI agent, the agent means 5341 performs the process of passing the generated information to the AI ​​agent.

[0788] (Step S4513) The agent means 5341 decides whether or not to terminate the dialogue. If it decides to terminate the dialogue, it returns to the higher-level process; otherwise, it proceeds to step S4511. The dialogue is terminated, for example, when the above-mentioned dialogue termination conditions are met.

[0789] (Step S4514) The agent means 5341 determines whether or not it has received dialogue information, which is information from one or more dialogue partners. If dialogue information has been received, the process proceeds to step S4515; otherwise, it returns to step S4514.

[0790] (Step S4515) The agent means 5341 stores the one or more dialogue information received in step S4514 in the claim management unit 511, associating the claim identifier with the identifier of the dialogue partner. Here, it is preferable for the agent means 5341 to obtain the time from a clock (not shown) and store the time as associated with the dialogue information.

[0791] (Step S4516) Agent means 5341 passes one or more conversational pieces of information to the third-party AI agent. Return to step S4206.

[0792] In addition, in the flowchart of Figure 45, if no dialogue information is received from the dialogue partner for a predetermined period of time or longer in step S4514, the dialogue process may be terminated and the process may return to a higher level.

[0793] Furthermore, in the flowchart of Figure 45, there may be two or more parties with whom the third-party AI agent interacts. Two or more parties with whom the interaction is conducted include, for example, the petitioner, the respondent, one or more other third parties, the petitioner's AI agent, the respondent's AI agent, and one or more third-party AI agents.

[0794] Next, an example of third-party support processing in step S3628 will be explained using the flowchart in Figure 46.

[0795] (Step S4601) The agent means 5341 obtains a claim identifier that identifies the claim in question.

[0796] (Step S4602) The agent means 5341 determines whether or not there is an agent type identifier "Third-Party Support AI Agent" that corresponds to the claim identifier obtained in step S4601, based on the agent management information paired with the claim identifier in the claim management unit 511. If a Third-Party Support AI Agent exists, the process proceeds to step S4604; otherwise, the process proceeds to step S4603.

[0797] (Step S4603) Agent means 5341 performs third-party support AI agent configuration processing.

[0798] An example of the third-party support AI agent configuration process will be explained using the flowchart in Figure 47. In step S4603, the agent means 5341 passes an agent type identifier that identifies the third-party support AI agent to the module that performs the AI ​​agent configuration process.

[0799] (Step S4604) The agent means 5341 obtains one or more pieces of information from the claim management unit 511 that are paired with the claim identifier obtained in step S4601.

[0800] (Step S4605) The agent means 5341 passes one or more pieces of information obtained in step S4604 to the third-party support AI agent.

[0801] (Step S4606) Agent means 5341 determines whether or not it has obtained a response from the third-party support AI agent. If a response has been obtained, it proceeds to step S4607; otherwise, it returns to step S4606.

[0802] (Step S4607) Step S4007) The agent means 5341 obtains generated information from the response obtained in step S4606 and stores it in association with the claim iden...

Claims

1. A negotiation support device that assists in at least some phases of a dispute between a petitioner and a respondent, from the commencement of negotiations to the conclusion of negotiations or the completion of the performance of the agreed-upon terms, A prompt management unit that stores one or more AI agent configuration prompts from among: a petitioner AI agent configuration prompt, which is a system prompt for configuring a petitioner AI agent, which is an AI agent that assists the petitioner in the negotiations on behalf of the petitioner; a respondent AI agent configuration prompt, which is a system prompt for configuring a respondent AI agent, which is an AI agent that assists the respondent in the negotiations on behalf of the respondent; a petitioner support AI agent configuration prompt, which is a system prompt for configuring a petitioner support AI agent, which is an AI agent that assists the petitioner; a respondent support AI agent configuration prompt, which is a system prompt for configuring a respondent support AI agent, which is an AI agent that assists the respondent; and a third-party support AI agent configuration prompt, which is a system prompt for configuring a third-party support AI agent, which is an AI agent that assists a third party in the negotiations. A petition acquisition unit that acquires petition information having a petitioner identifier that identifies the petitioner making the petition, a respondent identifier that identifies the respondent, and petition content information relating to the content of the said petition, A petition storage unit that stores the aforementioned petition information, The system comprises a negotiation support department that assists in negotiations between the petitioner and the respondent in the petition identified by the petition information, The aforementioned Negotiation Support Department, The process involves providing the aforementioned petitioner AI agent configuration prompt to the generating AI, configuring the petitioner AI agent, providing the petitioner AI agent with the petitioner information, and performing petitioner proxy processing, which is the process of interacting with the respondent or the third party or other AI agents on behalf of the petitioner, or The respondent AI agent configuration prompt is given to the generating AI, the respondent AI agent is configured, the respondent AI agent is given the claim information, and the respondent agent performs respondent proxy processing, which is the process of interacting with the claimant or the third party or other AI agents on behalf of the respondent, or The process involves providing the aforementioned petitioner support AI agent configuration prompt to the generating AI, configuring the petitioner support AI agent, providing the petition information to the petitioner support AI agent, obtaining a response from the petitioner support AI agent, and outputting the response to the petitioner, or The respondent support processing is performed by providing the respondent support AI agent configuration prompt to the generating AI, configuring the respondent support AI agent, providing the respondent support AI agent with the claim information, obtaining a response from the respondent support AI agent, and outputting the response to the respondent, or A negotiation support device comprising agent means for performing third-party support processing, which involves providing the aforementioned third-party support AI agent configuration prompt to a generating AI, configuring the aforementioned third-party support AI agent, providing the aforementioned claim information to the aforementioned third-party support AI agent, obtaining a response from the aforementioned third-party support AI agent, and outputting the response to the aforementioned third party.

2. The prompt management unit includes: It stores settlement proposal prompts, which are prompts that include instructions to create a settlement proposal for a dispute. The agent means is The negotiation support device according to claim 1, comprising a settlement proposal means that, in the negotiation, provides a generating AI with one or more types of information from among negotiation information entered by the applicant, negotiation information entered by the respondent, negotiation information entered by the third party, or generated information which is information output by any of the AI ​​agents, and a settlement proposal creation prompt, obtains a settlement proposal from the generating AI, and outputs the settlement proposal to one or more participants from among the applicant, the respondent, or the third party.

3. A negotiation support device that supports at least some of the phases of a dispute between a petitioner and a respondent, from the commencement of negotiations to the conclusion of negotiations or the completion of performance of the agreed-upon terms, A settlement proposal prompt is a prompt that includes instructions to create a settlement proposal for a dispute, A prompt management unit that stores one or more AI agent configuration prompts from the following: a petitioner AI agent configuration prompt, which is a system prompt for configuring a petitioner AI agent, which is an AI agent that assists the petitioner in the negotiations on behalf of the petitioner; a respondent AI agent configuration prompt, which is a system prompt for configuring a respondent AI agent, which is an AI agent that assists the respondent in the negotiations on behalf of the respondent; a third-party AI agent configuration prompt, which is a system prompt for configuring a third-party AI agent, which is an AI agent that assists the negotiations on behalf of the third party assisting the negotiations; a petitioner support AI agent configuration prompt, which is a system prompt for configuring a petitioner support AI agent, which is an AI agent that assists the petitioner; a respondent support AI agent configuration prompt, which is a system prompt for configuring a respondent support AI agent, which is an AI agent that assists the respondent; and a third-party support AI agent configuration prompt, which is a system prompt for configuring a third-party support AI agent, which is an AI agent that assists the third party assisting the negotiations. A petition acquisition unit that acquires petition information having a petitioner identifier that identifies the petitioner making the petition, a respondent identifier that identifies the respondent, and petition content information relating to the content of the said petition, A petition storage unit that stores the aforementioned petition information, The system comprises a negotiation support department that assists in negotiations between the petitioner and the respondent in the petition identified by the petition information, The aforementioned Negotiation Support Department, The process involves providing the aforementioned petitioner AI agent configuration prompt to the generating AI, configuring the petitioner AI agent, providing the petitioner AI agent with the petitioner information, and performing petitioner proxy processing, which is the process of interacting with the respondent or the third party or other AI agents on behalf of the petitioner, or The respondent AI agent configuration prompt is given to the generating AI, the respondent AI agent is configured, the respondent AI agent is given the claim information, and the respondent agent performs respondent proxy processing, which is the process of interacting with the claimant or the third party or other AI agents on behalf of the respondent, or The generating AI is given the aforementioned third-party AI agent configuration prompt, the third-party AI agent is configured, the third-party AI agent is given the aforementioned claim information, and the third-party proxy processing is performed, which is the process of interacting with the claimant or the respondent or other AI agents on behalf of the third party, or The process involves providing the aforementioned petitioner support AI agent configuration prompt to the generating AI, configuring the petitioner support AI agent, providing the petition information to the petitioner support AI agent, obtaining a response from the petitioner support AI agent, and outputting the response to the petitioner, or The respondent support processing is performed by providing the respondent support AI agent configuration prompt to the generating AI, configuring the respondent support AI agent, providing the respondent support AI agent with the claim information, obtaining a response from the respondent support AI agent, and outputting the response to the respondent, or The system comprises agent means for performing third-party support processing, which involves providing the aforementioned third-party support AI agent configuration prompt to the generating AI, configuring the aforementioned third-party support AI agent, providing the aforementioned claim information to the aforementioned third-party support AI agent, obtaining a response from the aforementioned third-party support AI agent, and outputting the response to the aforementioned third party. The agent means is In the aforementioned negotiations, the system provides a generating AI with one or more types of information, including negotiation information entered by the applicant, negotiation information entered by the respondent, negotiation information entered by the third party, or generated information output by any of the AI ​​agents, as well as a settlement proposal creation prompt, obtains a settlement proposal from the generating AI, and outputs the settlement proposal to one or more of the parties involved, including the applicant, the respondent, or the third party, and includes a means for proposing a settlement proposal. The aforementioned means of proposing a settlement proposal is, A negotiation support device that retrieves the application status information paired with the application identifier of a said application from an application management unit, which stores application status information that is information containing application information and is paired with an application identifier that identifies one or more applications, and outputs a settlement proposal if the said application status information, or information based on said application status information, satisfies the conditions for creating a settlement proposal.

4. The prompt management unit includes: The aforementioned petitioner AI agent configuration prompt is stored, The agent means is After performing the aforementioned processing on behalf of the petitioner, The aforementioned processing on behalf of the petitioner is, The negotiation support device according to claim 1, which includes a proxy processing that receives negotiation information from the respondent or the third party, or acquires dialogue information or generated information output from an AI agent other than the petitioner's AI agent, passes the negotiation information or the dialogue information or the generated information to the petitioner's AI agent, obtains a response from the petitioner's AI agent, and outputs negotiation information based on the response to the respondent or the third party or an AI agent other than the petitioner's AI agent.

5. The agent means is The negotiation support device according to claim 4, which performs the applicant's proxy processing when the applicant agent cooperation conditions are met, such as receiving an instruction to act on behalf of the applicant, receiving negotiation information transmitted from the respondent, or the class of the negotiation information transmitted from the respondent being a specific class.

6. The prompt management unit includes: The respondent's AI agent configuration prompt is stored there, The agent means is Performing the aforementioned representative processing for the respondent, The aforementioned handling of the respondent's behalf is as follows: The negotiation support device according to claim 1, which includes a proxy processing step of receiving negotiation information from the petitioner or the third party, or acquiring dialogue information or generated information output from an AI agent other than the respondent AI agent, passing the negotiation information or the dialogue information or the generated information to the respondent AI agent, obtaining a response from the respondent AI agent, and outputting negotiation information based on the response to the petitioner, the third party, or an AI agent other than the respondent AI agent.

7. The agent means is The negotiation support device according to claim 6, which performs the respondent proxy processing when the respondent agent cooperation conditions are met, such as receiving a proxy instruction from the respondent, receiving negotiation information transmitted from the petitioner, or the class of the negotiation information transmitted from the petitioner being a specific class.

8. The prompt management unit includes: The system stores one or more different third-party AI agent configuration prompts, which are system prompts for configuring a third-party AI agent that is an AI agent that assists in the negotiations on behalf of the third party assisting in the negotiations. The agent means is The generating AI is given one or more different third-party AI agent configuration prompts, the claim information is given to each of the one or more third-party AI agents, and third-party proxy processing is performed, which is the process of communicating with one or two of the parties involved, either the claimant or the respondent, on behalf of each of the one or more third parties. The aforementioned third-party processing is, The negotiation support device according to claim 1, which includes a proxy processing that receives negotiation information from the petitioner or the respondent, or acquires dialogue information or generated information output from an AI agent other than the third-party AI agent, passes the negotiation information or the dialogue information or the generated information to the third-party AI agent, obtains a response from the third-party AI agent, and outputs negotiation information based on the response to an AI agent other than the petitioner or the respondent or the third-party AI agent.

9. The agent means is The negotiation support device according to claim 8, which performs the third-party proxy processing when the conditions for third-party agent cooperation are met, such as receiving a proxy instruction from a third party, receiving negotiation information transmitted by the petitioner or respondent, or the class of the negotiation information transmitted by the petitioner or respondent being a specific class.

10. The negotiation support device according to claim 8, wherein the one or more different third-party AI agents are one or more AI agents selected from among an AI agent acting on behalf of a mediator, an AI agent acting on behalf of an arbitrator, an AI agent acting on behalf of a supporter, or an AI agent acting on behalf of the secretariat.

11. The prompt management unit includes: The aforementioned petitioner support AI agent configuration prompt is stored, The agent means is The negotiation support device according to claim 1, which provides the aforementioned petitioner support AI agent configuration prompt to a generating AI, configures the petitioner support AI agent, provides the petitioner support AI agent with the petitioner support AI agent with the petitioner support AI agent with the petitioner support AI agent with the petitioner support AI agent with the petitioner support AI agent with the petitioner support AI agent with the petitioner support AI agent with the petitioner support AI agent with the petitioner support AI agent with the petitioner support AI agent, obtains a response from the petitioner support AI agent and outputs the response to the petitioner.

12. The agent means is The negotiation support device according to claim 11, which performs the petitioner support processing when the petitioner support agent cooperation conditions are met, such as receiving a support instruction from the petitioner, receiving negotiation information from the respondent, or the class of the negotiation information received from the respondent being a specific class.

13. The prompt management unit includes: The aforementioned respondent support AI agent configuration prompt is stored there, The agent means is The negotiation support device according to claim 1, which provides the respondent support AI agent configuration prompt to the generating AI, configures the respondent support AI agent, provides the respondent support AI agent with the claim information, provides the respondent support AI agent with negotiation information received during negotiations, or dialogue information indicating the content of conversations between the participants or the AI ​​agent and the AI ​​agent during negotiations, or generated information output by the AI ​​agent during negotiations, obtains a response from the respondent support AI agent, and outputs the response to the respondent.

14. The agent means is The negotiation support device according to claim 13, which performs the respondent support processing when the respondent support agent cooperation conditions are met, such as receiving a support instruction from the respondent, receiving negotiation information from the petitioner, or the class of the negotiation information received from the petitioner being a specific class.

15. The prompt management unit includes: The aforementioned third-party support AI agent configuration prompt is stored, The agent means is The negotiation support device according to claim 1, which provides the generating AI with the third-party support AI agent configuration prompt, configures the third-party support AI agent, provides the third-party support AI agent with the claim information, provides the third-party support AI agent with negotiation information received during negotiations, or dialogue information indicating the content of conversations between the participants or the AI ​​agent and the AI ​​agent during negotiations, or generated information output by the AI ​​agent during negotiations, obtains a response from the third-party support AI agent, and outputs the response to the third party.

16. The agent means is The negotiation support device according to claim 15, which performs the third-party support processing when the conditions for collaboration with a third-party support agent are met, namely that support instructions have been received from a third party, negotiation information has been received from the petitioner or respondent, and the class of the negotiation information received from the petitioner or respondent is a specific class.

17. The negotiation support device according to claim 15, wherein the one or more different third-party support AI agents are one or more AI agents selected from among a mediator support AI agent, an arbitrator support AI agent, an assistant support AI agent, or an administrative office support AI agent.

18. The prompt management unit includes: The system contains a neutrality judgment AI agent configuration prompt, which is a system prompt for configuring a neutrality judgment AI agent that performs a neutrality assessment regarding negotiations, obtains a neutrality assessment result, and outputs the result of the neutrality assessment. The agent means is Negotiation support device according to claim 1, comprising: providing the neutrality judgment AI agent configuration prompt to a generating AI; providing the neutrality judgment AI agent with one or two types of information from the claim information and generated information which is negotiation information or dialogue information entered by the claimant, or negotiation information or dialogue information entered by the respondent, or negotiation information or dialogue information entered by a third party, or information output by the AI ​​agent; obtaining the neutrality evaluation result from the neutrality judgment AI agent; and outputting the neutrality evaluation result or information based on the neutrality evaluation result.

19. The neutrality evaluation means is, The negotiation support device according to claim 18, which outputs the neutrality evaluation result or information based on the neutrality evaluation result when the neutrality evaluation result satisfies the neutrality output conditions.

20. The agent means is The negotiation support device according to claim 1, which obtains one or more pieces of information from a storage unit that stores a collection of static information such as laws and regulations, articles, amendment history, court precedents, case law commentaries, legal literature, articles, academic theories, forms, procedural information, industry information, industry-specific legal risks, industry practices, administrative documents, comparative legal information, overseas legal systems, or a legal term dictionary, or a collection of dynamic information such as application information, consultation content, party input content, chat history, voice transcript, telephone log, submitted materials, evidence, party comments, mediator comments, secretariat comments, AI proposal history, mediator selection / modification / rejection logs, settlement proposals, settlement / failure results, performance status, recovery results, participation rate, response rate, and period to resolution, using the application information, negotiation information received during the negotiation, dialogue information indicating the content of conversations between an involved party or another AI agent and the AI ​​agent during the negotiation, or generated information output by another AI agent during the negotiation, and passes the one or more pieces of information to any one or more of the AI ​​agents.

21. A negotiation support method that causes a computer to perform all processing performed by the negotiation support device described in any one of claims 1 to 20.

22. Computers, A program for causing a negotiation support device to function as described in any one of claims 1 to 20.

Citation Information

Patent Citations

  • A system and method for providing dispute resolution for electronic payment transactions.

    JP2015535365A

  • Cort dispute settlement procedure system

    JP2020184245A

  • System

    JP2025053546A

  • System

    JP2026029376A

  • Negotiation support device, negotiation support method, and program

    JP2026047100A