Legal affairs consultation support system

Abstract

To smoothly solve a problem related to legal affairs in an enterprise by reducing a consultation work burden on a person in charge of the legal affairs and providing a quick and accurate answer to an employee.SOLUTION: A legal affairs consultation support system includes: a storage unit for storing a prompt for generating answer information to consultation information; a consultation information acquisition unit for acquiring the consultation information from a requester; an assignment unit for assigning the acquired consultation information and the prompt to a generation AI; and an answer provision unit for providing the generated answer information to a person in charge of the legal affairs. By being constituted by this manner, it is achieved to reduce time required to respond to an inquiry about a legal affairs consultation and improve the quality of the inquiry, thereby contributing to increase in efficiency of work relating to legal affairs in a company.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a legal consultation support system that enables legal personnel to efficiently and quickly answer questions from employees and the like. [Background technology]

[0002] As a digital transformation (DX) technology in the legal field, technologies that support contract review work have been disclosed (for example, Patent Documents 1 to 3). On the other hand, a technology for facilitating smooth communication between legal personnel and contract personnel (on-site personnel) has also been disclosed (Patent Document 4). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2020-013524 [Patent Document 2] Japanese Patent Publication No. 2021-174474 [Patent Document 3] Patent Publication No. 2021-117844 [Patent Document 4] Japanese Patent Application Publication No. 2020-166870 Summary of the Invention [Problem to be solved by the invention]

[0004] However, legal staff currently manually answer each and every legal issue that arises within a company. This process is time-consuming and places a heavy burden on legal staff. Furthermore, if employees repeatedly ask the same questions, they often have to create the same answers each time.

[0005] The present invention aims to reduce these inefficiencies and ease the workload of legal personnel, and by providing employees with fast and accurate answers, it also facilitates the resolution of legal issues within a company. [Means for solving the problem]

[0006] According to the present invention, legal personnel can use generation AI to automatically generate answers, significantly reducing the time and effort required to create answers. Furthermore, by using prompts stored in the memory unit, it is possible to provide consistent, high-quality answers. Furthermore, by quickly providing answers to employees through the answer provider, the speed of response to legal issues across the entire company is improved. This is expected to improve the quality of risk management for the company and increase the satisfaction of legal personnel and employees.

[0007] According to the present invention, a storage unit that stores a prompt for generating response information to the consultation information; a consultation information acquisition unit that acquires the consultation information acquired from a client; an assigning unit that assigns the consultation information and the prompt to a generative AI; and an answer providing unit that provides the answer information generated by the generative AI to a legal officer. A legal consultation support system is available. [Effects of the Invention]

[0008] According to the present invention, the consultation workload of legal affairs personnel is reduced, and by providing employees with prompt and accurate answers, legal issues within a company can be resolved smoothly. [Brief explanation of the drawings]

[0009] [Figure 1] 1 is a diagram showing an example of the configuration of a legal consultation support system (hereinafter referred to as "system") according to an embodiment of the present invention. [Figure 2]FIG. 2 is a functional block diagram of the legal consultation support server of FIG. 1. [Figure 3] 10 is an example of the structure of consultation information. [Figure 4] 2 is a sequence diagram showing the flow of processing in the legal consultation support system of FIG. 1. FIG. [Figure 5] 1. FIG. 4 is another functional block diagram of the legal consultation support server of FIG. [Figure 6] 1. FIG. 4 is another sequence diagram showing the flow of processing in the legal consultation support system of FIG. DETAILED DESCRIPTION OF THE INVENTION

[0010] The present invention will be described below by listing the contents of the embodiments. The present invention has the following configuration. [Item 1] a storage unit that stores a prompt for generating response information to the consultation information; a consultation information acquisition unit that acquires the consultation information acquired from a client; an assigning unit that assigns the consultation information and the prompt to a generative AI; and an answer providing unit that provides the answer information generated by the generative AI to a legal officer. Legal consultation support system. [Item 2] further comprising a correction receiving unit that receives corrections to the generated answer information. Item 1. The legal consultation support system. [Item 3] the consultation information includes at least contract information and question information, the prompt includes instructions to generate a summary of the agreement information. Item 1. The legal consultation support system. [Item 4] The contract information is commented contract information including comment information from the legal staff, The assigning unit assigns the commented contract information to the generation system AI, The prompt includes a prompt for generating a response to the requester based on the contract information with comments. Item 3. The legal consultation support system. [Item 5] The contract information is commented contract information including comment information from the legal staff, The assigning unit assigns the commented contract information to the generation system AI, The prompt includes a prompt for generating a response to the other party of the requester based on the contract information with comments. Item 3. The legal consultation support system. [Item 6] a response policy receiving unit for receiving a response policy from the legal affairs officer in response to the consultation information; The assigning unit includes the answer policy in the prompt. Item 1. The legal consultation support system. [Item 7] a generative AI usage acceptance unit that accepts a selection as to whether or not the consultation information and the prompt are to be assigned to the generative AI by the assignment unit; a response receiving unit that, when the selection of not granting the right is received, prompts the legal affairs officer to respond to the client, Item 1. The legal consultation support system. [Item 8] a communication acquisition unit that acquires communication information between at least the legal affairs officer and other legal affairs officers in response to the consultation information; The assigning unit assigns the consultation information, the communication information, and the prompt to a generative AI. Item 1. The legal consultation support system. [Item 9] a storage unit for storing prompts for generating consultation information for legal personnel; a consultation information acquisition unit that acquires auxiliary information related to the consultation information acquired from a client; an assigning unit that assigns the auxiliary information and the prompt to the generative AI; a providing unit that provides the consultation information generated by the generation system AI to the client, Legal consultation support system. [Item 10] further comprising a correction receiving unit that receives corrections to the generated consultation information; Item 8. The legal consultation support system. [Item 11] the auxiliary information includes at least one of contract information and reference material information, the prompt includes instructions to generate a summary of the supplemental information. Item 8. The legal consultation support system. [Item 12] The prompt includes an instruction to indicate if the auxiliary information is insufficient. Item 8. The legal consultation support system. [Item 13] a storage unit that stores a prompt for generating response information to the consultation information; a consultation information acquisition unit that acquires the consultation information acquired from a client; an assigning unit that assigns the consultation information and the prompt to a generative AI; and an answer providing unit that provides the answer information generated by the generative AI to a legal officer. Legal consultation support server. [Item 14] Computer, a storage unit for storing a prompt for generating response information to the consultation information; a consultation information acquisition unit that acquires the consultation information acquired from the client; an assigning unit that assigns the consultation information and the prompt to the generative AI; The answer providing unit functions as an answer providing unit that provides the answer information generated by the generative AI to a legal officer. Legal consultation support program. [Item 15] On the computer, a storage step of storing a prompt for generating answer information to the consultation information; a consultation information acquisition step of acquiring the consultation information acquired from the client; an assigning step of assigning the consultation information and the prompt to a generative AI; an answer providing step of providing the answer information generated by the generative AI to a legal officer; Legal consultation support methods.

[0011] <Details of implementation form> Hereinafter, embodiments of the present invention will be described with reference to the drawings.

[0012] <Summary> The present invention relates to a legal consultation support system for legal consultations, and in particular to a system that enables legal personnel to efficiently and quickly generate and provide answers to questions from employees. This system automates the process of acquiring consultation information from employees and generating answer information based on that information using a generation AI, thereby reducing the workload of legal personnel and improving the efficiency of answer provision.

[0013] First Embodiment A specific use case of this invention can be seen in "responding to contract-related inquiries from employees." In this case, employees may ask questions about the interpretation of specific contract clauses, the application of a contract under specific circumstances, or how to negotiate with the other party. Traditionally, legal staff would have to respond to such questions individually and create answers while referring to relevant documents. This process is time-consuming and can lead to repeated inquiries.

[0014] The present invention also includes the use of generative AI to generate answers based on communication exchanges within the legal department, their history, and investigation results.

[0015] In the system according to this embodiment, an employee first inputs a question into the system through the consultation information acquisition unit. This question contains specific content, such as a question about a specific clause in a contract or the application of the contract in a specific business scenario. Next, the system's assignment unit combines this consultation information with a prompt (in this case, a prompt containing instructions for generating an answer based on legal knowledge related to the contract) previously prepared in the storage unit, and sends the combined information to the generation AI. Based on this information, the generation AI generates answer information including appropriate legal interpretations and guidelines. The generated answer information is provided to a legal staff member through the answer provision unit. The legal staff member reviews the answer information, makes adjustments or adds additional explanations as necessary, and then provides the answer to the employee. This process enables legal staff to provide quick and consistent answers to complex contract-related questions.

[0016] <Overall structure> As shown in Figure 1, a legal consultation support system (hereinafter sometimes simply referred to as the "system") 1 according to an embodiment of the present invention has a network configuration including a client terminal 10, a legal affairs officer terminal 20, a legal consultation support server 30, and a generative AI provision server 40.

[0017] The client terminal 10 is a terminal used by an employee and has the function of inputting and transmitting the consultation content. The legal affairs officer terminal 20 is a terminal used by a legal affairs department officer and has the function of checking the generated answer information and editing or approving it as necessary. The legal consultation support server 30 is a core server including a consultation information acquisition unit, an assignment unit, and an answer provision unit, and receives consultation information, transmits the information to the generative AI provision server 40, and provides answer information to the legal affairs officer. The generative AI provision server 40 is a server that generates answer information based on the consultation information using a generative AI.

[0018] These components are connected via the Internet or an in-house network. Consultation information from the client terminal 10 is sent to the legal affairs officer 20 using some kind of communication tool (an internal chat system, an email system, a dedicated application, etc.). The legal affairs officer terminal 20 then sends the consultation information to the legal consultation support server 30. The legal consultation support server 30 combines the consultation information with an appropriate prompt stored in a memory unit and sends the consultation information to the generative AI provision server 40. The generative AI provision server 40 generates answer information based on this information and sends it back to the legal consultation support server 30. The legal consultation support server 30 provides the received answer information to the legal affairs officer terminal 20. The legal affairs officer terminal 20 checks the answer information and provides the answer to the client terminal 10 after approving or editing it.

[0019] <Configuration of legal consultation support server> As shown in FIG. 2, the legal consultation support server 30 includes a storage unit 101, a consultation information acquisition unit 102, an assignment unit 103, a response provision unit 104, and a correction reception unit 105.

[0020] <Storage section> The storage unit 101 has a function for storing prompts used to generate response information to consultation information in the legal consultation support server 30. Prompts for generating response information to consultation information are stored in the storage unit 101. The prompts may include a wide range of information, such as extensive knowledge about the law to be referred to when providing a response, precedents, guidelines, industry standards, and interpretations of contract clauses.

[0021] Specifically, the prompt in the store may take the following form: (1) Keywords and phrases Examples of keywords and phrases that can help understand the content of a consultation and generate relevant answers include specific legal terms, contract clauses, and industry jargon. For example, if the consultation is about a confidentiality agreement, the system will generate an answer that includes these keywords. For labor-related consultations, keywords can also be included in prompts when providing guidance or interpretations based on the Labor Standards Act. (2) Instructions and Question Guidelines Examples of such prompts include specific instructions to guide responses to questions based on consultation information, and guidelines that indicate points to consider when creating a response. For example, when consulting about the interpretation of a specific clause in a contract, a prompt could be given to instruct the respondent to provide a detailed explanation of the clause's legal implications and its impact on the company. Furthermore, when legal precedents may affect the content of the consultation, a prompt could be given to provide an interpretation by referring to the latest legal precedents. (3) Precedents and case studies Prompts may contain information that shows specific cases, such as similar past consultations and their solutions, court cases, industry best practices, etc. For example, in response to a consultation about a breach of a confidentiality agreement, a prompt could be generated that generates an answer based on past court cases and their results. In addition, in response to a consultation about a dispute with an employee, a prompt could be displayed that shows a collection of case studies that suggest optimal responses that a company can take.

[0022] These prompts can be automatically selected or manually selected based on the content of the consultation, and can be combined appropriately to increase the accuracy and relevance of the response to the specific legal consultation. Furthermore, as described below, the legal staff may also incorporate response strategies into the prompts.

[0023] The memory unit 101 appropriately categorizes and indexes these prompts, enabling it to quickly search for the most relevant prompt based on the consultation content sent from the consultation information acquisition unit 102 and send it to the assignment unit 103. This search and selection process provides the basis for the generative AI to generate more accurate answers that respond appropriately to the target consultation content. In addition, the memory unit 101 may be dynamically updated, and new legal information, court cases, changes in corporate policies, etc. are added as prompts from time to time. This enables the system to always provide answers that reflect the latest legal knowledge.

[0024] As described above, the following process may be performed to search for a highly relevant prompt. First, the consultation content received by the consultation information acquisition unit 102 is analyzed using natural language processing (NLP). Through this analysis, important information, such as keywords, phrases, and related legal fields, is extracted from the consultation content. The extracted information is treated as metadata representative of the characteristics of the consultation content and serves as the basis for the search. Next, this metadata is used as a search query for a prompt database stored in the storage unit 101. The database in the storage unit 101 indexes prompts based on various characteristics (e.g., legal fields, keywords, frequency of use, etc.), enabling efficient searches. The search system compares the metadata of the consultation content with the prompt index to identify the most relevant prompt. The identified prompt is sent to the annotation unit 103. The annotation unit 103 combines this prompt with the consultation content to construct a final query for the generative AI. Based on this query, the generative AI generates answer information that appropriately corresponds to the specific consultation content.

[0025] <Consultation Information Acquisition Department> The consultation information acquisition unit 102 collects consultation details submitted by clients, such as company employees and other related parties, in the legal consultation support server 30 and converts them into a format that can be processed by the system. The consultation information acquisition unit 102 serves as an input interface in the system 1. Specifically, the consultation information acquisition unit receives consultation details submitted via text input, voice recognition, or a messaging platform such as email.

[0026] The received consultation content may be analyzed using natural language processing (NLP) technology to understand the gist of the consultation and extract important keywords and concepts. This analysis process extracts important information from the consultation content, which can then be stored and used as basic data for subsequent processing such as prompt search and answer generation.

[0027] The consultation information acquisition unit 102 may also convert consultation details submitted in various formats into a standardized predetermined data format. This maintains the flexibility and compatibility of the entire system, and allows clients using different input methods to easily use the system. The converted data may be combined with prompts managed in the storage unit 101 and sent to the generative AI via the assignment unit.

[0028] <attachment part> In the legal consultation support server 30, the assignment unit 103 combines the consultation content provided by the consultation information acquisition unit with the related prompts stored in the storage unit 101 to finally form a query for the generative AI. Through an appropriate combination of the consultation content and the prompt, the assignment unit 103 enables the generative AI to generate the most appropriate and accurate answer for a specific consultation.

[0029] The assignment unit 103 integrates the specific consultation content from the client with a prompt prepared in advance for responding to the legal consultation. The integration of the prompts may be performed manually by having a legal officer select the most appropriate prompt (or having a legal officer select from multiple prompts) based on the analysis results of the consultation content as described above, and then combines the prompt with the consultation content to construct a query in a format that is easy for the generative AI to process.

[0030] When constructing a query, the AI ​​must take into account the characteristics and context of the inquiry and adjust the prompt accordingly. For example, if a query asks for the interpretation of a specific legal term, a prompt containing the legal background and precedents related to that term will be selected and combined with the inquiry. This provides the AI ​​with enough information to generate a more detailed and specific answer.

[0031] The query finally formed by the assigning unit 103 is sent to the generative AI providing server 40, where the answer generation process by the AI ​​is executed.

[0032] <Answer provision department> The answer providing unit 104 provides the legal officer with the answer information generated by the generative AI. The answer providing unit 104 receives the output from the generative AI and provides an interface that allows the legal officer to check and process it. Through this interface, the legal officer can review the answer provided by the AI ​​and make corrections or add additional information as necessary. Finally, the answer approved by the legal officer is provided to the client.

[0033] <Correction Reception Department> Legal professionals may use their own expertise and experience to validate and appropriately modify the AI-generated answers, ensuring that the answers comply with actual legal, corporate policy, and practical requirements.

[0034] As shown in FIG. 2, the correction receiving unit 105 receives corrections from the legal staff to the generated answer information. The correction receiving unit 105 provides an interface that enables the legal staff to make corrections or add additional information to the answer information provided to the legal staff by the answer providing unit 104. The correction receiving unit 105 may include a text editor that enables the legal staff to directly edit the answer content, a function to add additional comments to the answer, a function to check the correction history, and the like. This allows the legal staff to provide an answer to the client after confirming that the answer is completely accurate and optimally addresses the client's question.

[0035] Furthermore, corrections made via the correction receiving unit 105 can be saved as feedback in the system 1 (not shown), which will enable the generative AI to generate more accurate answers in the future during its learning process. In other words, the correction receiving unit not only improves the quality of answers, but also contributes to expanding the knowledge base of the entire system and improving its accuracy.

[0036] <Consultation information> 3, the consultation information from the client in this embodiment may include contract information and question information. In this case, the prompt may include an instruction to generate a summary of the contract information (including at least the part of the contract that is the subject of the consultation).

[0037] One benefit of including contract information in the consultation information is that System 1 can directly analyze the content of the submitted contract and extract the information necessary to generate answers to questions based on that contract. The combination of contract information and question information provides a basis for generating more specific and relevant answers when the consultation relates to the interpretation or application of a specific contract. Furthermore, by including instructions to generate a summary of the contract information in the prompt, the generative AI begins by reading the entire contract and summarizing its key points and clauses. This summarization process allows the AI ​​to understand the core content of the contract and then formulate answers to specific questions. This is expected to reduce the labor required for legal personnel to generate prompts by eliminating the need for summarization.

[0038] Another benefit of including contract information in consultation information is that it also makes the legal professional's responses to the client easier to understand. This is because the contents of contracts and other legal documents are often difficult for non-experts to understand. By clearly and concisely presenting the key points and clauses of the contract through the summary, the client can gain a deeper understanding of the content of the consultation and the contract behind the question. The client can specifically understand which part of the contract their question relates to and which key points of that contract are relevant. This allows the client to consider their own situation in light of the content of the contract and more easily grasp the logic behind the answer provided by the legal professional.

[0039] <Response Policy Reception Department> As shown in FIG. 2, the system according to this embodiment may further include a response policy receiving unit 106. In this case, the assigning unit 103 may be configured to include this response policy in the prompt. The response policy receiving unit 106 is a component that enables legal professionals to provide response policies that determine the direction and focus of responses to consultation content. The response policy received through this unit is incorporated into the prompt by the assigning unit 103 and is taken into consideration in the response generation process by the generative AI.

[0040] The response policy receiving unit 106 allows legal personnel to specify in advance specific perspectives, interpretations, and key points to emphasize in their responses to inquiries. For example, legal personnel can instruct the generative AI on which response policy to use when multiple response policies are possible, such as emphasizing a specific risk when multiple risks exist, emphasizing a specific legal principle, or prioritizing multiple benefits when multiple benefits exist. This ensures that the generated response is in line with the legal personnel's intentions. The assignment unit 103 incorporates the received response policy as part of the prompt. This enables the generative AI to not only generate an answer based on the consultation content and prompt, but also to form a more specific and purposeful answer based on the response policy provided by the legal personnel. As a result, the generated answer not only directly responds to the client's question but also reflects the legal personnel's strategic perspective and advice, thereby increasing the value to the client.

[0041] <Response Policy Reception Department> 2, the communication acquisition unit 109 has a function of collecting, in addition to consultation information related to the client, communication information between legal staff regarding the consultation. This allows information regarding the content of the consultation to be acquired, such as the exchange of opinions between legal staff regarding the consultation, the progress of discussions, and the agreed-upon interpretation and response policy.

[0042] The assignment unit 103 integrates the acquired consultation information, communication information, and prompts, and assigns all of this information to the generative AI. This process enables the generative AI to generate answers by taking into account not only the consultation information but also a broader range of background information based on discussions and exchanges of opinions between legal professionals. As a result, the generated answer information can provide more specific information in response to the client's question. In this way, the addition of the communication acquisition unit enables the legal consultation support system to generate higher-quality answers by utilizing communications between legal professionals related to the consultation content.

[0043] <Generation AI> The generative AI utilized in this invention is a system designed based on natural language processing (NLP) technology and capable of generating relevant answers from text-based consultation information. This AI system learns from a large amount of legal documents, precedents, legal commentaries, corporate policy documents, etc. to build its knowledge base. The learning process uses deep learning and machine learning techniques to identify patterns that can be used to form appropriate answers for specific consultation content.

[0044] Generative AI receives consultation information and prompts related to that information and automatically generates answers containing legal interpretations and advice for specific issues. This process involves forming answers through multiple steps, including analyzing the consultation content, applying relevant legal principles, and comparing with previous similar cases. The AI ​​system may have self-learning capabilities. It continuously improves its accuracy and efficiency by receiving feedback on its answers from legal professionals. Feedback from legal professionals, including corrections and additional instructions, is used as learning data for the AI ​​and serves as the basis for generating more accurate and useful answers in the future.

[0045] <Processing flow> The flow of the system 1 according to this embodiment will be described with reference to FIG. 4. The client terminal 10 transmits consultation information to the legal affairs officer terminal 20 (SQ001). The legal affairs officer terminal 20 logs in to the legal consultation support system 1 and transmits the received consultation information (SQ002 and SQ003). The collected consultation information, including specific questions and information about the contract in question, is analyzed using natural language processing technology as necessary to extract important keywords and concepts. Next, the legal consultation support server 30 selects the most relevant prompt from its storage unit based on the consultation content (SQ004). This prompt may include instructions for generating an answer or guidelines regarding a specific legal issue. The selected prompt is assigned to the generative AI via the assigning unit (SQ005). The generative AI generates appropriate answer information based on the prompt and the consultation information, and transmits it to the legal consultation support server 30 (SQ006). The generated answer information is provided to the legal officer through the answer providing unit (SQ007). The legal officer reviews this answer and may make corrections or add additional information as necessary. After the legal officer has made any necessary corrections, the answer is finally provided to the client (SQ008).

[0046] <Second embodiment> In the legal consultation support system of the present invention, the contract information may be commented contract information that includes comment information from legal personnel (comment information added to the contract document, revision history information, etc.). The assigning unit 103 may assign this commented contract information to the generation system AI.

[0047] Furthermore, the prompt may include instructions for generating a response to the client based on the commented contract information, particularly the comment information. For example, a contract document may contain numerous comment information added by legal staff. If the prompt includes instructions to explain, summarize, or list the comment information, the creation of a response to the client can be made more efficient. In this case, by summarizing comments added on the same date, comments added by the same editor, or comments added in the same version of the document based on information such as the date the comment was added, the editor who added the comment, or the version information of the document to which the comment was added, a response to the client can be generated based on information only about the updated parts.

[0048] <Third embodiment> While the system 1 according to the second embodiment described above is designed to generate a response from a legal professional to a client, it may also be designed to suggest a message that the client should send to a counterparty, in addition to or separately from the response to the client. In this case, the prompt includes instructions for generating a response to the client's counterparty based on the commented contract information. This additional element allows the system to provide more precise and practical advice for specific situations and problems that the client may encounter in negotiations or interactions with counterparties.

[0049] <Fourth embodiment> System 1 according to this embodiment may further include generative AI usage acceptance unit 107 and answer acceptance unit 108. Generative AI usage acceptance unit 107 has a function of accepting from the legal officer a selection of whether or not to assign consultation information and prompts to the generative AI via assignment unit 103, and answer acceptance unit 108 provides an interface that enables the legal officer to provide an answer directly to the client when a selection is made not to use generative AI.

[0050] This configuration allows legal personnel to select the optimal response method for each consultation. Specifically, if the consultation is complex, if legal personnel want to provide individualized attention to a particular case, if particularly sensitive information is involved, or if there are concerns about data privacy and security, legal personnel can choose to skip the automated response generation process using generative AI and provide a direct response. This selection is made through the generative AI usage reception unit, and the response reception unit prompts the legal personnel to create a corresponding response based on this selection.

[0051] Even if a choice is made not to use generative AI, the answer receiving unit 108 may provide access to resources that are useful in creating an answer, such as legal documents related to the consultation content, case studies, answers to previous similar cases, etc. This allows legal personnel to efficiently create and provide answers to clients.

[0052] <Fifth embodiment> The systems according to the first to fourth embodiments described above are systems for legal personnel to generate and provide answers to questions from employees efficiently and quickly. However, as shown in Figure 5, the system according to the fifth embodiment of the present invention generates consultation information for a client to make to a legal personnel.

[0053] A specific use case of this invention can be seen in "responding to contract-related inquiries from employees." In this case, employees may ask questions about the interpretation of specific contract clauses, the application of a contract under specific circumstances, or how to negotiate with the other party. Traditionally, legal staff would have to respond to such questions individually and create answers while referring to relevant documents. This process is time-consuming and can lead to repeated inquiries.

[0054] In the system of the present invention, the client inputs auxiliary information related to the consultation content into the system. This auxiliary information is acquired by the auxiliary information acquisition unit 102. Examples of auxiliary information include information for more accurate analysis of the consultation information, such as the case category (contract review, legal consultation, general consultation, etc.), legal consultation type, attached contract, business terms, and case background. A prompt for appropriately responding to the consultation content is selected from the storage unit 101.

[0055] The selected prompt and the acquired auxiliary information are assigned to the generative AI via the assigning unit 103. The generative AI generates an answer based on the content of the consultation. This answer is provided to the requester via the providing unit 104a. In this embodiment, the client may modify the consultation information provided as needed, thereby ensuring that the consultation information matches the client's actual intentions and purposes.

[0056] As in the first embodiment, the generative AI may be configured to summarize the contents of the attached contract. This allows the legal staff to clearly and concisely present the important points and clauses of the contract, allowing them to gain a deeper understanding of the content of the contract that is the background to the consultation or question.

[0057] The prompt in this embodiment may include an instruction to "indicate if there is a lack of auxiliary information." This allows the generative AI to point out the lack of auxiliary information if it finds it is insufficient to generate an answer to the consultation content, and to prompt the requester to provide the necessary additional information. Based on the provided additional information, the generative AI can generate an answer again, and through this process, the accuracy of the answer provided is improved.

[0058] <Processing flow> The flow of the system 1 according to this embodiment will be described with reference to FIG. 6. The client terminal 10 logs in to the legal consultation support system 1 and transmits auxiliary information (SQ102 and SQ103). The collected consultation information, including specific questions and information about the contract in question, is analyzed using natural language processing technology as necessary to extract important keywords and concepts. Next, the legal consultation support server 30 selects the most relevant prompt from the storage unit based on the content of the auxiliary information (SQ104). This prompt may include instructions for generating the consultation information or guidelines regarding a specific legal issue. The selected prompt is transmitted to the generative AI via the transmission unit (SQ105). The generative AI generates appropriate consultation information based on the prompt and auxiliary information and transmits it to the legal consultation support server 30 (SQ106). The generated consultation information is provided to the client terminal 10 via the transmission unit (SQ107). The client may review the consultation information and make corrections or add additional information as necessary. After appropriate corrections have been made, the consultation information is finally provided to the legal officer (SQ108).

[0059] <Application example> The above-described system 1 is intended to mainly handle consultation information relating to contracts, but can also be suitably used for the following types of consultation, for example. (1) Labor law consultation in the human resources department In the human resources department, you will be able to provide prompt and appropriate legal advice to employees regarding working conditions, leave, dismissal, harassment, etc. (2) Regulatory compliance consultation in the compliance department We can provide advice based on the latest regulatory information regarding regulatory compliance in specific industries, such as finance, pharmaceuticals, and environmental protection. (3) Intellectual property rights consultation in the research and development department In the research and development department, we can provide specialized legal advice on patent applications, trademark registrations, and copyright protection for new inventions and technologies. (4) Contract management consultation in the legal department In the legal department, you may provide consultation regarding the drafting, review, and amendment of contracts, as well as advice on contract negotiation strategies. (5) Advertising law consultation in the marketing department We can provide appropriate guidelines and help you avoid legal risks when consulting about legal restrictions on marketing activities and advertising laws.

[0060] As described above, the system of the present invention has the potential to improve the efficiency, speed, and accuracy of providing information services in a variety of fields.

[0061] <Hardware configuration example> The systems are configured to be able to communicate with each other via the Internet. This system may be configured as a cloud / network type system provided by a designated business operator, or as an on-premise system operated independently within the company that introduces the system.

[0062] Each of the above-mentioned functional blocks can be configured by, for example, hardware provided in a server device (terminal device), a DSP (Digital Signal Processor), or software. For example, when configured by software, each of the above-mentioned functional blocks is actually configured with a CPU, RAM, ROM, etc. of a computer, and is realized by the operation of a program stored in a recording medium such as RAM, ROM, a hard disk, or a semiconductor memory.

[0063] The above-described embodiment is merely an example for facilitating understanding of the present invention, and is not intended to limit the present invention. The present invention can be modified and improved without departing from the spirit thereof, and it goes without saying that the present invention includes equivalents thereof. [Explanation of symbols]

[0064] 10 Legal Affairs Officer Terminal 20. Client terminal 100 Legal consultation support server

Claims

1. a storage unit for storing a prompt for generating response information to consultation information from a client; a consultation information acquisition unit that acquires the consultation information acquired from the client; an assigning unit that assigns the consultation information and the prompt to a generative AI; An answer providing unit that provides the answer information generated by the generation system AI to a legal officer, Legal consultation support system.

2. further comprising a correction receiving unit that receives corrections to the generated answer information. The legal consultation support system according to claim 1.

3. the consultation information includes at least contract information and question information, the prompt includes instructions to generate a summary of the agreement information. The legal consultation support system according to claim 1.

4. The contract information is commented contract information including comment information from the legal staff, The assigning unit assigns the commented contract information to the generation system AI, The prompt includes a prompt for generating a response to the requester based on the contract information with comments. The legal consultation support system according to claim 3.

5. The contract information is commented contract information including comment information from the legal staff, The assigning unit assigns the commented contract information to the generation system AI, The prompt includes a prompt for generating a response to the other party of the requester based on the contract information with comments. The legal consultation support system according to claim 3.

6. a response policy receiving unit for receiving a response policy from the legal affairs officer in response to the consultation information; The assigning unit includes the answer policy in the prompt. The legal consultation support system according to claim 1.

7. a generative AI usage reception unit that receives a selection of whether or not the provision unit should provide the consultation information and the prompt to the generative AI; a response receiving unit that, when the selection of not granting the right is received, prompts the legal affairs officer to respond to the client, The legal consultation support system according to claim 1.

8. a communication acquisition unit that acquires communication information between at least the legal affairs officer and other legal affairs officers in response to the consultation information; The assigning unit assigns the consultation information, the communication information, and the prompt to a generative AI. The legal consultation support system according to claim 1.

9. a storage unit for storing prompts for generating consultation information for legal personnel; a consultation information acquisition unit that acquires auxiliary information related to the consultation information acquired from a client; an assigning unit that assigns the auxiliary information and the prompt to the generation system AI; a providing unit that provides the consultation information generated by the generation system AI to the client, Legal consultation support system.

10. further comprising a correction receiving unit that receives corrections to the generated consultation information; The legal consultation support system according to claim 8.

11. the auxiliary information includes at least one of contract information and reference material information, the prompt includes instructions to generate a summary of the supplemental information. The legal consultation support system according to claim 8.

12. The prompt includes an instruction to indicate if the auxiliary information is insufficient. The legal consultation support system according to claim 8.

13. a storage unit that stores a prompt for generating response information to the consultation information; a consultation information acquisition unit that acquires the consultation information acquired from a client; an assigning unit that assigns the consultation information and the prompt to a generative AI; An answer providing unit that provides the answer information generated by the generation system AI to a legal officer, Legal consultation support server.

14. Computer, a storage unit for storing a prompt for generating response information to the consultation information; a consultation information acquisition unit that acquires the consultation information acquired from the client; an assigning unit that assigns the consultation information and the prompt to the generation system AI; The system functions as an answer providing unit that provides the answer information generated by the generation system AI to a legal officer. Legal consultation support program.

15. On the computer, a storage step of storing a prompt for generating answer information to the consultation information; a consultation information acquisition step of acquiring the consultation information acquired from the client; an assignment step of assigning the consultation information and the prompt to a generative AI; an answer providing step of providing the answer information generated by the generation system AI to a legal officer; Legal consultation support methods.

Images