AI character status assurance system, program and method

Abstract

The AI ​​character's state changes are monitored and predicted in real time, ensuring that the AI ​​character is in an inexperienced state. [Solution] In an AI character provision system that guarantees a specific state of an AI character, the state parameter management unit 36a of the character management unit 36 ​​provided in the AI ​​character management server 3 includes an inexperienced state setting unit 361 that sets initial values ​​of state parameters that form the characteristics of the AI ​​character, including at least one of personality, experience, or knowledge, as inexperienced state data, a fluctuation prediction unit 362 that monitors the input and output of the AI ​​character and predicts fluctuations in the state parameters, a history recording unit 366 that periodically records the current values ​​and predicted fluctuations of the state parameters as change history on the blockchain network, and a state assurance unit 368 that analyzes the change history recorded by the history recording unit 366 and calculates its identity with respect to the inexperienced state data.

Description

[Technical Field]

[0001] The present invention relates to a system that guarantees a specific state of an AI character, such as an initial state or a state adjusted for a specific user, and in particular to an AI character state guarantee system, program, and method that sets the AI ​​character's characteristics, such as personality, experience, and knowledge, as the initial state and prevents these characteristics from fluctuating within a game or simulation. [Background technology]

[0002] In recent years, advances in AI (artificial intelligence) have made it possible for AI to have natural voice dialogue with humans, and its applications are expanding beyond everyday conversation to include child education and elderly care. Because AI can be a conversation partner, it is desirable to set up a personality and establish a dialogue based on the range of knowledge and virtual experience appropriate to that role, under the setting of a virtual character such as a school teacher, friend, parent, or sibling.

[0003] In particular, when setting the personality of an AI as the opposite sex in a dating simulation, there is sometimes value in maintaining the character's initial state, without any experience, learning, or growth.

[0004] To maintain the initial state of data, a mechanism is required to ensure that data has not been tampered with, and one such mechanism is the security management device disclosed in Patent Document 1. The system disclosed in Patent Document 1 creates or updates the items of managed files based on a whitelist for approval applications, and outputs the items of managed files in the created or updated whitelist as file changes, thereby providing a security management device that reduces the work costs of approval applicants and approvers in procedures aimed at preventing fraud in computer production processes or computer adjustment processes. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Japanese Patent Application Publication No. 2023-173481 Summary of the Invention [Problem to be solved by the invention]

[0006] The neural network model underlying an AI character constantly changes depending on user interactions and the passage of time. However, the security management device disclosed in the aforementioned Patent Document 1 monitors only static records of file changes, and is therefore unable to fully manage the status of parameters such as the personality, experience, and knowledge of an AI character based on a neural network model that constantly changes.

[0007] Therefore, the present invention aims to solve the above problems and provide an AI character state assurance system, program, and method that monitors and predicts changes in the state of an AI character in real time and guarantees the initial state of the AI ​​character. [Means for solving the problem]

[0008] In order to solve the above problem, the present invention is a system that guarantees an inexperienced state of an AI character, an inexperienced state setting unit that sets initial values ​​of state parameters forming characteristics of the AI ​​character, including at least one of personality, experience, and knowledge, as inexperienced state data; a fluctuation prediction unit that monitors input and output of the AI ​​character and predicts fluctuations in the state parameters; a history recording unit that periodically records the current values ​​of the state parameters and the predicted fluctuations as change history on a blockchain network; a state assurance unit that analyzes the change history recorded by the history recording unit and calculates identity for the unexperienced state data; The present invention is characterized by comprising:

[0009] The present invention also provides a method for causing a computer to maintain an inexperienced state of an AI character, comprising: an inexperienced state setting step in which an inexperienced state setting unit sets initial values ​​of state parameters forming characteristics of the AI ​​character, including at least one of personality, experience, and knowledge, as inexperienced state data; a fluctuation prediction step in which a fluctuation prediction unit monitors input and output of the AI ​​character and predicts fluctuations in the state parameters; a history recording step in which a history recording unit periodically records the current values ​​of the state parameters and the predicted fluctuations as change history on a blockchain network; a state assurance step in which a state assurance unit analyzes the change history recorded by the history recording unit and calculates identity with respect to the unexperienced state data; The present invention is characterized by comprising:

[0010] In the above invention, a comparison unit that compares a value of a trigger parameter included in the state parameter with a threshold value that causes a specific event to occur; an event execution unit that executes the specific event in response to a comparison result by the comparison unit; Furthermore, The history recording unit adds to the change history whether or not the specific event has occurred and the number of times the specific event has occurred. It is preferable.

[0011] In the above invention, a knowledge database stores knowledge restriction information for restricting a knowledge range corresponding to current state parameters, and also stores keywords or commands that form prompts for the AI ​​character; a change rate control unit that varies a change rate of the state parameter in response to the occurrence of the specific event; a query generation unit that extracts keywords or commands from a knowledge database based on the current value of the state parameter and the rate of change of the state parameter that has been changed in response to the occurrence of the specific event; an information request control unit that generates an information request that requests an information response from the AI ​​character using the keyword or command extracted by the query generation unit; a filtering unit that compares the response content from the AI ​​character with the knowledge restriction information and corrects the response content that deviates from the knowledge range; It is preferable that the device further comprises:

[0012] In the above invention, the specific event is: Only the first user who executed the next specific event is permitted to execute the next specific event and have a conversation about it, and subsequent users who are subsequent users are restricted from executing the next specific event and having a conversation about it; The AI ​​character's knowledge scope is updated only through the interaction with the first user. It is preferable.

[0013] The above-described system and method according to the present invention can be realized by executing a program of the present invention written in a predetermined language on a computer. That is, by installing the program of the present invention in an IC chip or memory device of a portable terminal device, smartphone, wearable terminal, mobile PC or other information processing terminal, or a general-purpose computer such as a personal computer or server computer and executing it on a CPU, a system having the above-described functions can be constructed and the method according to the present invention can be implemented.

[0014] Furthermore, the program of the present invention can be distributed, for example, via a communication line, and can be transferred as a package application that runs on a stand-alone computer by recording it on a computer-readable recording medium. Specifically, this recording medium can be recorded on a variety of recording media, including magnetic recording media such as flexible disks and cassette tapes, optical disks such as CD-ROMs and DVD-ROMs, and RAM cards. Furthermore, a computer-readable recording medium on which this program is recorded makes it possible to easily implement the above-described system and method using a general-purpose computer or a dedicated computer, and also makes it easy to store, transport, and install the program. [Effects of the Invention]

[0015] As described above, according to the present invention, by setting initial values ​​for state parameters that form the AI ​​character's characteristics such as personality, experience, and knowledge, and using the blockchain mechanism to strictly manage the history of changes to these settings, an innocent, inexperienced state that is not influenced by others is guaranteed, allowing the user to enjoy a realistic first experience with the character.

[0016] In addition, the system monitors input and output to the AI ​​character in real time, predicting fluctuations in state parameters to determine whether the character has simply learned various phenomena as knowledge or whether it has gained real-life experience that will influence its subsequent behavior and values, and thereby managing the change history of inexperienced state data.By recording the change history on the blockchain network, the change history can be maintained in an immutable form, preventing data tampering and ensuring the authenticity of the character.

[0017] In particular, by analyzing the recorded change history and calculating the identity with respect to the virgin state data, it is possible to evaluate how much the character has changed from the initial setting. Therefore, even if there is a change history, by calculating the identity between the virgin state and the current state, it is possible to evaluate and guarantee whether an effective virgin state is maintained, thereby preventing unintended changes to the character's characteristics and allowing the user to continue to enjoy a consistent character experience.

[0018] As a result, according to the present invention, it is possible to monitor and predict changes in the state of an AI character in real time, and ensure that the AI ​​character is in an inexperienced state. [Brief explanation of the drawings]

[0019] [Figure 1] 1 is a conceptual diagram showing the overall configuration of an AI character providing system according to an embodiment. FIG. [Figure 2] 1A is a block diagram showing the internal configuration of an AI character management server according to an embodiment, and FIG. 1B is a block diagram showing the internal configuration of a character control unit. [Figure 3] FIG. 2A is a block diagram showing the internal configuration of a smartphone according to an embodiment, and FIG. 2B is a block diagram showing the configuration of an interface control unit. [Figure 4] 1 is a block diagram showing an internal configuration of a guarantee system according to an embodiment. [Figure 5] FIG. 1 is a flow chart showing the operation of the AI ​​character provision service according to the embodiment. [Figure 6] FIG. 10 is a sequence diagram showing an operation related to event execution according to the embodiment. [Figure 7] FIG. 1 is an explanatory diagram of a blockchain in an AI character provision system according to an embodiment. [Figure 8] FIG. 10 is an explanatory diagram regarding the occurrence of an event in the AI ​​character provision service according to the embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0020] Below, with reference to the accompanying drawings, embodiments of the image rights management method, management system, and management program according to the present invention will be described in detail. In this embodiment, an example is given of using the environmental value digital art generation method according to the present invention in an energy trading service provided by an AI character provision system. Note that the embodiments shown below are merely examples of devices embodying the technical concept of the present invention, and the technical concept of the present invention does not limit the material, shape, structure, arrangement, etc. of each component to those described below. The technical concept of the present invention may be modified in various ways within the scope of the claims.

[0021] (Overview of the AI ​​character provision service) In this embodiment, an AI character provision service is implemented by an AI character provision system. Figure 1 shows the overall configuration of the AI ​​character provision service according to this embodiment. In this embodiment, the AI ​​character status assurance system, method, and program of the present invention are implemented in an AI character provision system 1 constructed on a communication network 2.

[0022] This AI character service allows users to experience various interpersonal simulations through AI characters provided using NLP (natural language processing) technology. For example, it can provide a romance simulation in which the AI ​​character is a virtual member of the opposite sex, an educational simulation in which the AI ​​character is a virtual teacher or student, a business simulation in which the AI ​​character is a virtual boss or customer, a medical simulation or counseling simulation in which the AI ​​character is a virtual medical professional or nurse, or an intercultural communication simulation in which the AI ​​character is a virtual foreigner.

[0023] Here we present an example of applying this to a dating simulation in which the AI ​​character is a virtual woman. In this dating simulation, the user deepens their relationship with the AI ​​character through dialogue and progresses the story by clearing various events. The AI ​​character responds to the user's input in real time, and the character's personality and feelings toward the user change, developing a deeper relationship with the user as the story unfolds.

[0024] This interpersonal simulation uses an NLP engine to analyze user input and generate natural dialogue, responding to user voice and text input in real time. Initial settings are configured for this AI character at the start of the simulation, including the AI ​​character's personality, appearance, background story, etc., and in particular, the opposite-sex experience points and romantic experience points related to experience events that will occur and progress during the simulation are set to 0.

[0025] In particular, the initial settings of this AI character can be customized to reflect the user's own profile, for example, by having the AI ​​character read and analyze the user's actual profile, giving the AI ​​character a profile that includes anecdotes such as "a childhood friend who knew the user from childhood," and then creating an inexperienced state with zero experience with the opposite sex and romance. Images, videos, and email sending and receiving history uploaded and stored by the user on cloud storage are analyzed, and the character is customized to a specific user's specifications based on the extracted user information.

[0026] In this dating simulation, the story progresses through dialogue with characters and events. For example, certain keywords or actions spoken by the user can increase the user's affinity parameter with the AI ​​character or affect their romantic feelings parameter, triggering the next experience event, and clearing it will allow the player to progress to the next stage.

[0027] This AI character learns through interactions with the user and through experiential events, and its responses change depending on changes in the relationship. By filtering queries when retrieving information from the knowledge database and answers obtained from NLP, responses are constrained according to the AI ​​character's experience level and stage. Furthermore, NLP technology is used to predict and recognize the user's emotions and reflect them in the AI ​​character's responses. For example, if the user appears sad, the character can offer words of encouragement. Knowledge data acquired from user interactions is continually added to this knowledge database, and the input of this knowledge data is recorded as a change history. By referencing this, responses based on past conversations can be based on user feedback, allowing the AI ​​character's responses to change and grow from moment to moment.

[0028] The story progresses through events that occur when the user's favorability parameter or the AI ​​character's experience points or emotion parameters exceed a certain threshold, as well as through periodic events. As the story progresses, knowledge data is added to the knowledge database, and filters are updated for queries used to retrieve information from the knowledge database and answers obtained from NLP. Restricted responses are gradually lifted according to the AI ​​character's new experience points and stage.

[0029] The experiential events mentioned above are romantic experiences that AI characters acquire in dating simulations, and these experiences influence the values, ways of thinking, and behavior they attain. Specifically, in addition to standard seasonal events such as birthdays and Christmas, there are also individual event scenarios based on the character's likeability toward the user, and relationships with characters change through special conversations and events. In particular, there are also random and sudden negative events such as troubles and mishaps, and problem-solving scenarios in which the user solves problems the character is facing. The user's advice and actions to solve these problems affect the AI ​​character's trust and likeability, which in turn determines the next event that occurs and the stage that progresses.

[0030] (Overall configuration of the AI ​​character provision system) Specifically, in this embodiment, a user receives voice data from a smartphone 5 used by the user, and receives an AI character provision service managed and operated by an AI character management server 3. More specifically, the AI ​​character provision system 1 according to this embodiment is configured by interconnecting an AI character management server 3, which executes processing to implement an AI character, and a smartphone 5, which is an information processing terminal used by each user to display and output the AI ​​character, via a communication network 2. In this embodiment, the smartphone 5 is described as an example of an information processing terminal. Also, a guarantee system 4 is installed on the communication network 2, which provides a blockchain interface service that guarantees the provision of the AI ​​character, and the guarantee system 4 allows connection to a distributed ledger system such as a blockchain.

[0031] The communication network 2 is an IP network such as the Internet that uses the communication protocol TCP / IP, and is a distributed communication network constructed by interconnecting various communication lines (public lines such as telephone lines, ISDN lines, ADSL lines, and optical lines, dedicated lines, fourth-generation (4G) communication methods such as LTE, fifth-generation (5G) and later communication methods, and wireless communication networks such as Wifi (registered trademark) and Bluetooth (registered trademark). This IP network also includes LANs such as intranets (corporate networks) and home networks using 10BASE-T, 100BASE-TX, etc.

[0032] The AI ​​character management server 3 is a server device that manages and operates AI characters. In this embodiment, it has the function of displaying and outputting AI characters via smartphone 5 and can be implemented as a single server device or a group of multiple server devices. The AI ​​character management server 3 also virtually implements multiple functional modules on a CPU, with each functional module cooperating to execute processing. The AI ​​character management server 3 can send and receive data via the communications network 2 using its communications function, and can present websites and provide services via browser software using its web server function.

[0033] On the other hand, the smartphone 5 is a portable information processing terminal device that uses wireless communication, and communicates wirelessly with a relay point such as a wireless base station 22, allowing the user to receive communication services such as phone calls and data communications while moving. Examples of communication methods include 3G (3rd Generation), LTE (Long Term Evolution), 4G, FDMA, TDMA, CDMA, W-CDMA, and PHS (Personal Handyphone System). The smartphone 5 is also equipped with various functions, such as a digital camera, an application software execution function, and a location information acquisition function using a GPS (Global Positioning System), and also includes mobile computers such as tablet PCs.

[0034] The location information acquisition function of the smartphone 5 is a function to acquire and record location information indicating the location of the smartphone itself. Examples of this location information acquisition function include a method of detecting the location of the smartphone itself using signals from a satellite 21, such as GPS, and a method of detecting the location using radio wave strength from a mobile phone wireless base station 22 or a Wi-Fi communication access point.

[0035] The smartphone 5 includes a display unit 53a such as a liquid crystal display that displays information, a voice input device for user interaction, and operation devices such as operation buttons for input operations, and the operation devices include a touch panel 52a that is superimposed on the liquid crystal display and serves as an input unit that acquires operation signals by touch operations such as specifying coordinate positions on the liquid crystal display. Specifically, the touch panel 52a is an input device that inputs operation signals by pressure or electrostatic detection by touch operations using the user's fingertip or pen, and is configured by superimposing a liquid crystal display that displays graphics and a touch sensor that receives operation signals corresponding to the coordinate positions of the graphics displayed on the liquid crystal display.

[0036] (Internal structure of each device) Next, we will explain the internal structure of each device that makes up the above-mentioned AI character provision system 1. Figure 2 shows the internal structure of the AI ​​character management server 3 according to this embodiment, and Figure 3 shows the internal structure of the smartphone 5 according to this embodiment. Note that the term "module" used in this explanation refers to a functional unit that is composed of hardware such as a device or equipment, software with such functionality, or a combination of these, and that performs a predetermined operation.

[0037] (1) AI character management server 3 First, we will explain the internal configuration of the AI ​​character management server 3. The AI ​​character management server 3 is a server device located on the communication network 2, and is capable of sending and receiving data to and from each smartphone 5 via the communication network 2.

[0038] Specifically, the AI ​​character management server 3 includes a communication interface 31 that communicates data via the communication network 2, an authentication unit 33 that authenticates the authority of users and user terminals, a location information management unit 32 that collects and manages location information of each user terminal, a character control unit 36 ​​that executes control related to the AI ​​character, an information distribution unit 34 that distributes information to each user, and various database groups 35a to 35c.

[0039] The database group includes a user database 35a that stores information about users, a character management database 35b, and an event management database 35c. Each of these databases may be a single database, or may be divided into multiple databases and a relational database in which each piece of data is linked together by setting relationships between them.

[0040] The information stored in the user database 35a includes an identifier (user ID, terminal ID) that identifies the user or the information processing terminal used by the user, and authentication information linked to a password, etc., as well as personal information of the user linked to the user ID and the model of the terminal device.

[0041] The character management database 35b is a storage unit that manages and stores the characteristics of AI characters, and its main state parameters include personality (e.g., introversion, logical), experience (e.g., event / mission completion, skill acquisition), and knowledge (conversational knowledge, general knowledge, specialized knowledge, etc.). The character's growth history, behavioral history, appearance (e.g., hair color, clothing), backstory (e.g., upbringing, past events), relationships (e.g., friends, romantic relationships), skills and abilities (e.g., academic ability, physical strength, occupational skills, etc.), and emotional state (e.g., happiness, sadness) are also managed as state parameters. Furthermore, the character management database 35b has a hierarchical structure for efficiently managing a character's basic information, personality, experience, knowledge, relationships, skills, and emotions, and provides a consistent and rich user experience by managing the character's complex characteristics in detail.

[0042] In this embodiment, the character management database 35b includes: Character Table: Manage basic information for each character Personality table: Detailed management of personality traits Experience Table: Manage each character's experience as a history Knowledge Table: Manages the knowledge information held by the character. Relationship Table: Manage relationships with other characters Skills table: Manages information about skills and abilities Emotion Table: Manage the emotional state of your character These data tables form a data set that can be referenced by mutual relationships.

[0043] The event management database 35c is an accumulation unit that manages and stores detailed experience events to enhance the growth of AI characters and the user experience. Key information includes an event ID that uniquely identifies each event, an event name that indicates the name and content of the event, and an event description. Additionally, the event type (e.g., date, work, education), the trigger conditions for the event, and the rewards and results obtained upon event completion are recorded. The database also manages the characters associated with the event, the event occurrence date and time, the character status parameters that change depending on the event (e.g., favorability, skill level), and the event progress status (in progress, completed, failed). This allows for event history management and consistent control of character growth and changes.

[0044] The authentication unit 33 is a module that establishes a communication session with each smartphone 5 via the communication interface 31 and performs authentication processing for each established communication session. This authentication processing involves obtaining authentication information from the smartphone 5 of the accessing user, referencing the user database 35a to identify the user, and authenticating their authority. The authentication results (user ID, authentication time, session ID, etc.) by the authentication unit 33 are sent to the character control unit 36 ​​and are also stored in the user database 35a as an authentication history.

[0045] The location information management unit 32 is a module that acquires location information acquired on the user's smartphone 5 and sent to the AI ​​character management server 3. The location information management unit 32 links the identifiers (user ID, terminal ID, etc.) of the user or user terminal device identified by the authentication process by the authentication unit 33 with the location information and stores it as usage history in the user database 35a.

[0046] The character control unit 36 ​​is a module that generates an AI character and controls its behavior while displaying and outputting the AI ​​character in cooperation with an application on the smartphone 5. In this embodiment, the character control unit 36 ​​functions as a fluctuation prediction unit that monitors input and output to and from the AI ​​character and predicts fluctuations in state parameters, and also functions as a history recording unit that records the current values ​​and predicted fluctuations of state parameters for each character as a change history on the blockchain network via the blockchain linking unit 36d periodically or as needed.

[0047] Specifically, the character control unit 36 ​​cooperates with the NLP collaboration unit 36b to update the blockchain to manage the execution history and state parameter update history of each AI character. Furthermore, specifically, the character control unit 36 ​​synchronizes with apps and data on the user's device, updating data such as the user's progress, character status, and dialogue history in real time to maintain consistency. The character control unit 36 ​​then records the state parameters in the character management database 35b and updates the database whenever the state parameters change.

[0048] In this embodiment, the character control unit 36 ​​includes a state parameter management unit 36a, an NLP collaboration unit 36b, an AI character state assurance unit 36c, and a blockchain collaboration unit 36d. The state parameter management unit 36a is a module that manages and controls various state parameters of the AI ​​character customized for each user. These state parameters include the following:

[0049] Personality: extroverted or introverted, conscientious or not, curious or not, logical or not, etc. Experience: This includes the number of missions (experience events) cleared, experience points, and skill usage experience. Depending on this experience points, knowledge and values ​​(priorities) will change as your knowledge and experiences expand. Knowledge: Conversational knowledge, general knowledge, specialized knowledge, etc., and the breadth and depth of knowledge is determined according to age and experience. - Character growth history and behavior history Appearance: This includes physical characteristics such as hair and skin color, as well as parameters that change depending on clothing, fashion sense, health, fatigue level, etc. It also includes the types of accessories worn, as well as stories about items such as favorites and mementos. Backstory: Includes character development based on upbringing, family background, past events, and relationships related to your profile. Relationships: This includes relationships with users, such as friendships, romantic relationships, family relationships, and work relationships, as well as geographical and physical relationships based on preferences such as hometown and residence. Skills and abilities: academic ability, physical strength, professional skills, Emotional state: Includes parameters such as happiness, anxiety, and stress, and depending on the degree and combination of these, alertness, affinity, and liking for the user are affected.

[0050] Specifically, the state parameter management unit 36a has an inexperienced state setting unit 361, a fluctuation prediction unit 362, and a change rate control unit 363. The inexperienced state setting unit 361 is a module that sets initial values ​​of state parameters that form the characteristics of the AI ​​character, including at least one of personality, experience, and knowledge, as inexperienced state data. On the other hand, the fluctuation prediction unit 362 is a module that monitors the input and output of the AI ​​character and predicts fluctuations in the state parameters, and the change rate control unit 363 is a module that changes the change rate of the state parameters in response to the occurrence of a specific event.

[0051] The state parameter management unit 36a realizes the following functions using the inexperienced state setting unit 361, the fluctuation prediction unit 362, and the change rate control unit 363. For example, the fluctuation prediction unit 362 has a function for managing experiential events for each AI character, and predicts or calculates changes in the above-mentioned state parameters according to the experiential events that have been cleared, and reflects these changes in the character management database 35b. The experiential event management function of this fluctuation prediction unit 362 monitors the occurrence conditions of various experiential events, executes triggered events, tracks the progress, analyzes the results, and reflects them in the state parameters. By tracking the progress and results of events, the experiential event management function changes the skills and abilities of the AI ​​character, reflecting them in interactions with the user.

[0052] The NLP linking unit 36b is a module that executes natural language processing (NLP) to analyze input from the user and generate an appropriate response. The NLP linking unit 36b cooperates with an external NLP service to establish a dialogue with the user. During this dialogue, in order to generate a response that corresponds to the characteristics (age, gender, personality, experience, knowledge, etc.) of the AI ​​character realized by NLP, the NLP linking unit 36b performs processes such as referencing a knowledge database based on the character's experience level, generating an information request, verifying the response, outputting information, and reflecting the dialogue results in the database, as described below.

[0053] Query generation and execution function The NLP collaboration unit 36b generates and executes a query to reference the character management database 35b, which is a knowledge database, based on the experience level of the character, and causes the NLP to respond with a range of knowledge corresponding to the status, level, and stage based on each character's experience value. Specifically, the NLP collaboration unit 36b acquires the experience value level of the AI ​​character by referencing the experience value table of the character management database 35b, and generates a query based on the experience level while referencing the filter management database to obtain knowledge within the knowledge range corresponding to the experience value level, and executes the query against the knowledge table.

[0054] Information request generation function The NLP service analyzes the results of the query execution and generates and sends an information request to accurately obtain information in response to the user's question or dialogue. This information request includes prompts, other commands, and negative prompts. The external NLP service receives this information request, generates a response, and sends it back to the NLP collaboration unit 36b.

[0055] * NLP response check function The response from the NLP is checked to ensure its accuracy and appropriateness, and if there are no problems, it is output as is, but if there are problems, the query or information request is made again as necessary. Information output function Next, the NLP linking unit 36b outputs information to the user through the AI ​​character. Based on the content of the NLP response, the character's words, actions, and behavior are determined, and the character conveys the information to the user. At this time, the AI ​​character's current emotional and physical state is reflected in the character's facial expressions and behavior. 〇Database reflection function The NLP linking unit 36b then reflects the knowledge and experience acquired as a result of the dialogue in the database. In particular, the newly learned knowledge, emotions, and experiences are reflected in the character management database 35b, which is a state parameter management database. The newly learned experiences are also recorded in the database as background information and past events of the character.

[0056] The AI ​​character state assurance unit 36c is a module that sets initial values ​​of state parameters that form the characteristics of the AI ​​character, including at least one of personality, experience, and knowledge, as inexperienced state data. The AI ​​character state assurance unit 36c according to this embodiment includes a comparison unit 364, an inexperienced state change unit 365, a history recording unit 366, an event execution unit 367, and a state assurance unit 368.

[0057] The comparison unit 364 is a module that compares the value of a trigger parameter included in the state parameter with a threshold value that causes a specific event to occur. The inexperienced state change unit 365 is a module that causes the specific event to occur according to the comparison result by the comparison unit.

[0058] The history recording unit 366 is a module that periodically records the current values ​​of the state parameters and the predicted fluctuations as a change history on the blockchain network. The history recording unit 366 also has a function of adding to the change history whether or not a specific event has occurred and the number of times it has occurred. The event execution unit 367 is a module that executes the occurrence of the specific event according to the comparison result by the comparison unit. The state assurance unit 368 is a module that analyzes the change history recorded by the history recording unit and calculates the identity with respect to the unexperienced state data.

[0059] In addition, the AI ​​character state assurance unit 36c has a function of analyzing the change history recorded by the blockchain cooperation unit 36d as a history recording unit and calculating the identity with respect to the inexperienced state data. The blockchain collaboration unit 36d is a module that requests the assurance system 4 on the network to manage the state parameters of the AI ​​characters used by each user as well as perform security management and other processes necessary for providing the AI ​​characters, and works in cooperation with the assurance system 4 to carry out the processes. The character control unit 36 ​​manages the AI ​​characters used by each user by collaborating with the assurance system 4 via the blockchain collaboration unit 36d.

[0060] Here, the assurance system 4, which is a blockchain interface service, has multiple nodes that encrypt and store at least a portion of the information related to the AI ​​character. These nodes aggregate at least a portion of the information related to the AI ​​character at predetermined times to form blocks, link these blocks to existing blocks to form a blockchain, and share the blockchain among multiple nodes and store it as a distributed ledger.

[0061] The information distribution unit 34 is a module that, in response to access from users, presents a website related to the AI ​​character management service, provides each user with a unique personal page, and allows each user to individually view the display output of the AI ​​character they own and information about the AI ​​character.

[0062] (2) Smartphone 5 Next, we will explain the internal configuration of the smartphone 5. As shown in Figures 3(a) and (b), the smartphone 5 is roughly composed of a communication interface 51, an input interface 52, an output interface 53, an application execution unit 54, and a memory 55.

[0063] The communication interface 51 is a communication device for making calls or data communications, and has the function of contactless communication such as wireless communication, and contact (wired) communication using a cable, adapter means, or the like. The input interface 52 is a device for inputting user operations, such as a mouse, keyboard, operation buttons, or touch panel 52a. The output interface 53 is a device for outputting video and audio, such as a display or speaker. In particular, the output interface 53 includes a display unit 53a such as a liquid crystal display, which is superimposed on the touch panel 52a, which is the input interface.

[0064] Memory 55 is a storage device that stores an OS (Operating System), firmware, various application programs, other data, etc., and this memory 55 stores a user ID that identifies the user, application data downloaded from the AI ​​character management server 3, and various data processed by the application execution unit 54. In particular, in this embodiment, memory 55 stores rights information, etc. obtained from the AI ​​character management server 3.

[0065] Application execution unit 54 is a module that executes applications such as a general OS, game application, browser software, etc., and is usually realized by a CPU, etc. In this application execution unit 54, an AI character control unit 543, a display data generation unit 546, a display control unit 545, and a position information acquisition unit 544 are virtually constructed by executing the program of the AI ​​character management application according to the present invention.

[0066] The AI ​​character control unit 543 is a module that controls an AI character on the smartphone 5 and establishes a dialogue with the user. In this embodiment, the AI ​​character control unit 543 includes an interface control unit 541, a graphics control unit 542, and a synchronization processing unit 540.

[0067] To realize the AI ​​character service on a smartphone app, the graphics control unit 542 uses computer graphics processing to generate and control 3D models and 2D avatars of the AI ​​characters, rendering the appearance and movements of the AI ​​characters in real time and displaying and outputting them on the user interface.

[0068] The graphics control unit 542 also controls the animation of the AI ​​character, such as appropriately changing the character's facial expressions and movements depending on the content of the dialogue with the user. At this time, it renders environmental graphics, such as the scene background, objects, and effects, in real time to create a sense of unity with the character. The graphics control unit 542 also generates graphics for interactive elements (buttons, menus, icons, etc.) operated by the user and dynamically displays them in response to touch operations. The graphics control unit 542 also plays related media content in response to dialogues and events, integrating, displaying, and outputting multimedia content, such as images, videos, and audio, to enhance the visual and auditory experience.

[0069] The interface control unit 541 is a module that enables smooth and intuitive interaction between the user (operator) and the AI ​​character, and manages the overall layout and design of the app, such as button placement on the application screen, menu configuration, and screen transitions, to provide an interface that is easy for the user to operate. The interface control unit 541 is also a module that processes user input (tap, swipe, voice input, etc.), acquiring and recognizing the user's touch operations and voice commands, and performing the corresponding actions.

[0070] Furthermore, the interface control unit 541 has a function to manage notifications and alerts for the user, such as displaying and outputting pop-up notifications and audio notifications when an event occurs or when a specific condition is met, as well as a function to configure a dialogue window for displaying dialogue with an AI character and to synchronize and save the user's operation history and setting information with the server. Specifically, as shown in FIG. 1(b), the interface control unit 541 has an input / output monitoring unit 541a, an experience level reference unit 541b, a knowledge database query generation unit 541c, and a knowledge database matching unit 541d.

[0071] The input / output monitoring unit 541a is a module that analyzes input from the user (text, voice, gestures, etc.) and extracts information for the AI ​​character to respond appropriately, and understands the user's intentions and emotions using NLP technology through the synchronization processing unit 540 and the NLP linkage unit 36b with the AI ​​character management server 3. The data analyzed by this input / output monitoring unit 541a is used to generate the character's response and for subsequent processing, categorizes the user's question, and records it in the character management database 35b on the AI ​​character management server 3 side as basic information included in the state parameters.

[0072] The experience level reference unit 541b is a module that references each of the databases 35a to 35d on the AI ​​character management server 3 to manage the current experience level of the AI ​​character, provides information used in interactions with the user and events, tracks the character's growth and learning progress, and references the corresponding experience points and levels. The experience level reference unit 541b appropriately changes the character's responses and actions according to its growth; for example, clearing a specific event increases the character's experience level, unlocking new skills and knowledge.

[0073] The knowledge database query generation unit 541c is a module that generates and executes queries to refer to the knowledge database based on the user's questions and requests. The knowledge database query generation unit 541c identifies the necessary information based on the data obtained from the input analysis unit, constructs a corresponding query, and sends it to the knowledge database on the AI ​​character management server 3, where it is used to obtain related information.

[0074] The knowledge database collator 541d is a module that analyzes information obtained from the knowledge database (in this embodiment, the knowledge table of the character management database 35b) based on the query sent by the query generator and generates an appropriate response. The knowledge database query generator 541c also collates the acquired data with the results of the input / output monitor 541a to confirm the accuracy and relevance of the information provided to the user. This process ensures that appropriate and useful information is provided to the user. The collation results are reflected in the AI ​​character's dialogue with the user.

[0075] Furthermore, the interface control unit 541 has an event condition determination unit 541e, an information request control unit 541f, a request filtering unit 541g, a response filtering unit 541h, and a state parameter management unit 541i. The event condition determination unit 541e is a module that monitors and determines the behavior of AI characters and the conditions for generating events. Specifically, the event condition determination unit 541e checks the conditions for generating specific events based on the user's behavior and the content of the dialogue, executes appropriate events in response to the user's interactions, and controls the character's behavior and story development. For example, it may generate a corresponding event when the user asks a specific question or uses a specific item.

[0076] The information request control unit 541f collects and processes information in response to a user request. Based on the information obtained from the input analysis unit, it generates an appropriate information request and sends it to the NLP. This allows it to obtain a response to the user from the NLP. The request filtering unit 541g is a module that filters the generated information requests and adjusts the request accordingly if the user's request exceeds the experience level or knowledge range of the AI ​​character so that appropriate information is provided.

[0077] The response filtering unit 541h is a module that filters information obtained from the knowledge database and responses from the NLP engine. Specifically, the response filtering unit 541h ensures that only information appropriate to the character's experience level and situation is provided to the user, and filters out information that is inappropriate or confusing to the user. For example, it ensures that responses to the user do not contradict the character's settings.

[0078] The state parameter management unit 541i is a module that manages the state parameters (personality, experience, knowledge, emotional state, etc.) of the AI ​​character, and appropriately updates parameters that fluctuate due to various events and interactions with the user, to reflect the character's growth and changes.

[0079] The synchronization processing unit 540 is a module that synchronizes the AI ​​character management processing on the smartphone 5 side with the AI ​​character management processing on the AI ​​character management server 3 side. Specifically, the AI ​​character management server 3 side predicts possible event processing based on the positions of other users' characters, the positions of objects, etc., generates the conditions for the event processing on the AI ​​character management server 3 side, transmits the conditions to the smartphone 5 side, and receives the conditions for the event processing. The actual generation of the event processing and the graphic processing for it are executed by the GUI control unit 545a on the smartphone 5 side based on the conditions for the event processing received from the AI ​​character management server 3. The results of the event processing executed by the GUI control unit 545a on the smartphone 5 side (such as the outcome of a battle or mini-game, or the score) are notified to the GUI control unit 545a on the AI ​​character management server 3 side via the synchronization processing unit 540, and are reflected in the subsequent game progress processing.

[0080] The display data generation unit 546 is a module that generates display data to be displayed on the display unit 53a. The display data is data generated by combining graphic data, image data, character data, video data, audio data, and other data, and generates, for example, a screen displaying a list of portrait rights or a detailed display of a specific AI character. The display process of the display data generated by the display data generation unit 546 is controlled by the display control unit 545.

[0081] The display control unit 545 is a module that controls specific screen display based on the data generated by the display data generation unit 546. In this embodiment, the display control unit 545 has a GUI control unit 545a, which realizes interaction with the user via the touch panel 52a and the display unit 53a, and performs appropriate screen transitions and information updates in response to user operations.

[0082] The GUI control unit 545a is a module in charge of displaying and controlling a graphical user interface (GUI) on the display of the smartphone 5. This GUI control unit 545a displays graphical elements such as various buttons, menus, icons, and windows on the display so that the user can intuitively operate applications, and performs appropriate reactions and actions in response to inputs such as touch operations from the user. The GUI control unit 545a operates in cooperation with the display data generation unit 546.

[0083] Based on the data and information generated by the display data generation unit 546, the GUI control unit 545a presents it to the user in a visually and easily operable format. For example, when displaying a list of voice actors, the GUI control unit 545a controls design elements such as the design, layout, and color of the list. Furthermore, the GUI control unit 545a also has the role of calling appropriate application functions in response to user operations. For example, when the user taps the "Record new voice data" button, the GUI control unit 545a receives the operation and sends a command to the AI ​​character control unit 543 to start the voice data recording process.

[0084] The location information acquisition unit 544 is a module having a function of acquiring current location information by using the built-in GPS sensor of the smartphone 5 and communication information with a wireless base station. Specifically, the location information acquisition unit 544 acquires the information by a Global Positioning System (GPS) using artificial satellites, base station positioning by triangulation based on the signal strength from the base station and base station information, Wi-Fi positioning using a database that combines the Wi-Fi SSID (Service Set ID) and signal conditions with longitude and latitude, etc.

[0085] In this embodiment, the location information acquisition unit 544 also includes a time information acquisition unit 544a. The location information and time information acquired by this location information acquisition unit 544 are linked as metadata, which is related data, when generating an AI character, and are saved as information on when and where the voice data was acquired. This function makes it possible to add temporal and geographical context to the voice data, thereby enriching the story and background behind the voice data.

[0086] (3) Guarantee System 4 As described above, in this embodiment, a guarantee system 4 is provided that guarantees copyright using blockchain technology. Specifically, as shown in Fig. 4, the guarantee system 4 includes a communication interface 43, an authentication unit 42, a rights management execution unit 44, an NFT transaction history database 41a, a key information database 41b, and an account database 41c.

[0087] The communication interface 43 is a module that transmits and receives data to and from other communication devices via the communication network 2, and in this embodiment, is connected to each AI character management server 3. The authentication unit 42 is a computer or software with that function that verifies the legitimacy of the accessing individual, and performs authentication processing based on the user ID that identifies each user. In this embodiment, the authentication unit 42 obtains the user's unique public address, public key, user ID, password, etc. from the accessing individual's smartphone 5 via the communication network 2, and verifies this against the key information database 41b to confirm whether the accessing individual has the right to access the site and whether the accessing individual is the individual in question.

[0088] The rights management execution unit 44 is a module that handles the AI ​​character's identity through NFT (Non-Fungible Token) technology. In this embodiment, the AI ​​character's identity is also a type of so-called digital asset, and the uniqueness and scarcity of the AI ​​character are verified and guaranteed through the rights management execution unit 44 based on blockchain technology. The rights management execution unit 44 uses this NFT technology to verify and guarantee the uniqueness and scarcity of the AI ​​character.

[0089] In this embodiment, an AI character created by a user is issued as an NFT, and related information and attributes are stored on the blockchain. Through this process, the originality and ownership of the AI ​​character are reliably protected, preventing unauthorized duplication or distribution by third parties. Furthermore, AI characters as NFTs can be traded on the market, allowing users to buy and sell rights related to their own voices as NFTs, creating new economic value.

[0090] Specifically, in this embodiment, the portrait rights in this NFT are linked to the public account of the legitimate owner, and by presenting the public account relationship, one can prove that one is the legitimate owner. Furthermore, only the legitimate owner can carry out transfer procedures such as assignment of the portrait rights.

[0091] The rights management execution unit 44 has a function to cooperate with the blockchain cooperation unit 36d on the AI ​​character management server 3 side, and this guarantee system cooperation function is requested by devices of other service providers, such as the blockchain cooperation unit 36d of the AI ​​character management server 3, to perform rights management processing such as security management for the provision of AI characters using NFTs, transaction records, and storage of service history, and proceeds with rights management processing via NFTs through interactions with each of these devices. Specifically, the rights management execution unit 44 includes a rights information provision unit 44a, a public address management unit 44b, a validity verification unit 44c, and a data update unit 44d.

[0092] The rights information providing unit 44a is a module that provides detailed rights information via the Internet when a user issues a cryptocurrency as an NFT. This provided information includes information such as the cryptocurrency price, the date of creation of the cryptocurrency (such as the user's age and generation), the current owner, and transfer benefits such as the number of times it has been traded.

[0093] The public address management unit 44b is a module that manages users' public accounts (public addresses) and has a private key issuance function that issues public addresses generated from public keys in public key cryptography to identify specific users, and private keys that can be paired with the public keys to identify the public keys and are used for digital signatures provided by AI characters via public addresses. These issued public addresses and their associated key information are stored in the key information database 41b. When a user trades NFTs, this public address serves as the key to verify the legitimacy of the transaction, and the public address functions as the user's identifier on the blockchain and is used to uniquely identify the NFTs owned by the user and their transaction history.

[0094] The legitimacy verification unit 44c is a module that verifies the legitimacy of transfers and transactions of NFTs. Specifically, it compares the public addresses managed by the public address management unit 44b with transaction information to verify that the transaction is being carried out by a legitimate owner. This prevents fraudulent transactions and unauthorized access by third parties. For example, the legitimacy verification unit 44c verifies that the NFTs involved in NFT transactions belong to the current owner through legitimate transactions and have not been tampered with. It also verifies the legitimacy of the NFTs using the current owner's public key linked to the public account. The legitimacy verification unit 44c also stores all transactions related to the NFTs in the NFT transaction history database 41a, and can verify the legitimacy of the transactions by comparing the public key with the NFT transaction history database 41a.

[0095] The data update unit 44d is a module that updates information related to the NFT. This module is activated when a user wishes to change the information related to the NFT or when a user acquires a new NFT. The data update unit adds new information to the blockchain and performs procedures to maintain consistency with previous information. For example, the data update unit 44d acquires information related to the NFT through the NFT, adds the public address of the new NFT owner, and changes the token owner verified in the distributed ledger, thereby transferring ownership of the NFT. Data updates by the data update unit 44d are protected by advanced security measures, providing robust protection against double transfers and tampering with transaction history.

[0096] (How to guarantee AI character status) The AI ​​character state assurance method of the present invention can be implemented by operating the AI ​​character provision system 1 described above. FIG. 5 is a sequence diagram showing the operation of the AI ​​character provision system. Note that the processing procedure described below is merely an example, and each process may be modified as much as possible. Furthermore, steps in the processing procedure described below may be omitted, replaced, or added as appropriate depending on the embodiment.

[0097] (1) Interaction behavior First, the user accesses the AI ​​character management server 3 using the smartphone 5, which serves as the user terminal, and after authentication processing, the service begins. Specifically, user U1 operates the AI ​​character dialogue application on smartphone 5 and attempts to access the AI ​​character management server 3. If the connection to the AI ​​character management server 3 is successful, user U1 is prompted to enter a user ID and password as login information. Once user U1 enters this information correctly, the server compares it with the authentication information stored in the user database 35a. If this authentication information matches, user U1 is successfully authenticated, user U1 is identified, and the AI ​​character provision service begins with settings specific to user U1.

[0098] Here, taking the case where a dialogue is initiated by user U1 on the user terminal side as an example, when user U1 performs an input operation such as voice input or text input through smartphone 5 (S101), interface control unit 541 analyzes the input (S102). Next, the current experience level of the character is referenced (S103). Specifically, experience level reference unit 541b references the experience value table in character management database 35b to obtain the experience level of the AI ​​character (S103).

[0099] Based on the acquired experience level, the knowledge database query generation unit 541c generates a query for the knowledge database and executes it against the character management database 35b (S104). This extracts information such as prompts and negative prompts to obtain responses within the range of knowledge corresponding to the AI ​​character's experience level (S105). At this time, the knowledge database comparison unit 541d compares the tags of the obtained information, and the request filtering unit 541g checks whether inappropriate information is included. Using the information whose tag has been checked in this way, the information request control unit 541f generates an information request for the NLP, and transmits the information request to the NLP from the synchronization processing unit 540 via the NLP linking unit 36b (S106).

[0100] In response to this information request, a response is returned from the NLP (S107). The response filtering unit 541h determines whether the content of this returned response is within the range of the knowledge level of the AI ​​character (S108). If the response is beyond the knowledge level ("N" in S108), steps S104 to S107 are repeated until the response content falls within the knowledge level. On the other hand, if the response content is within the range of the knowledge level ("N" in S108), the response is output to the user from the graphics control unit 542 (S109).

[0101] Then, the output response content is reflected in the knowledge database by the state parameter management unit 541i (S110), and the reflection result is recorded in the blockchain from the synchronization processing unit 540 via the blockchain linking unit 36d (S111).

[0102] (2) Actions when an event occurs Next, the operation when an event occurs in this embodiment will be described. In the AI ​​character provision service, the dialogue between the user and the NLP is constantly monitored (S201), fluctuations in state parameters such as the AI ​​character's knowledge level and its likeability to the user are predicted (S202), possible experiential events are searched for in the event management database 35c, the conditions for the occurrence of the event are compared with the current state parameters (S203), and the occurrence of the experiential event is determined (S204).

[0103] If the predicted value of the state parameter fluctuation does not exceed the occurrence condition ("N" in S204), steps S201 to S203 are repeated in a loop. On the other hand, if it is determined in step S204 that the predicted value of the state parameter fluctuation exceeds the occurrence condition ("Y" in S204), an event is generated (S205). This causes the AI ​​character's experience value to fluctuate, and the range of knowledge is opened up depending on the level of that experience value and the stage.

[0104] Execution of this event changes the experience points and knowledge level of the AI ​​character C1, changing state parameters such as experience points and knowledge level (S206). At this time, the inexperienced state is also updated. In response to the change in the state parameters, the rate of change of the character's state parameters thereafter is controlled (S207), and the control of this rate of change is recorded as history along with the change history of the state parameters (S208). This history recording is performed in cooperation with the blockchain platform (S209).

[0105] An example of a specific event in this embodiment will now be described. Figure 8 shows an overview of an experience event, which is one of the specific events that occurs in this embodiment. In this experience event, multiple users U1-U3 access a single AI character C1, and each can converse with them as one of many "friends" or "acquaintances." This state is a "general approachable state," in which all users are treated equally and fairly, with no special treatment given to anyone. For example, all second-person pronouns used in conversation are general, without proper names, such as "you" or "you." In this "general approachable state," each user can converse with a specific AI character equally, and status parameters such as favorability and intimacy with each user are added depending on the content of the conversation.

[0106] In the "wide approachable state," as users continue to converse with specific AI characters, status parameters such as favorability and intimacy increase, and the first user to clear the conditions for a specific event can then execute the next specific event. In this example, we will assume that user U1 executed the event first and achieved the status of "first user."

[0107] When user U1 completes a specific event, this specific AI character C1 transitions to a "limited approach state," in which only user U1 is permitted to execute the next specific event and interact with it. Additionally, in this limited approach state, subsequent users U2 and U3 become subsequent users, and users U2 and U3 are restricted from executing the next specific event and interacting with it.

[0108] In this limited approach state, the knowledge scope of the AI ​​character C1 is updated only through dialogue with the earliest user, user U1. In this "limited approach state," the AI ​​character C1 takes special measures to update its state parameters only for user U1, while for subsequent users U2 and U3, updates to their state parameters are restricted because user U1 is currently the earliest user. In this limited approach state, the second person pronoun used for user U1 is a proper noun, such as "U1-san," while the second person pronouns used for the other users U2 and U3 are still general pronouns, such as "you" or "you," without proper names.

[0109] In this "limited approach state," user U1 can have special conversations with AI character C1 that can affect state parameters, and state parameters such as favorability and intimacy toward user U1 are added depending on the content of the conversation.

[0110] (3) Operation when inexperienced state is set Here, we will explain the operation when setting, recording, and querying the inexperienced state of a specific AI character. Figure 7 is a flow diagram showing the operation when recording and querying the inexperienced state in the blockchain.

[0111] First, when a user creates a new AI character, an inexperienced state setting request is sent from the app on the smartphone 5 to the AI ​​character management server 3 (S301), and the inexperienced state setting process is executed on the AI ​​character management server 3 (S302).As a result of the execution, the generated inexperienced state parameters are recorded in each database, and the process of recording these various data on the blockchain is initiated.

[0112] Specifically, a process is executed to create a new account (here, a "public address for state assurance") on the blockchain through the assurance system 4 (S303), and data is saved (recorded) in this new "public address for state assurance" (S304). The recording process is completed by linking this "public address for state assurance" to the AI ​​character and starting management on the AI ​​character management server 3. Upon completion of creation of this "public address for state assurance" and linking it to an inexperienced AI character, a completion notification is sent to the user (S305), and the new AI character becomes accessible through the information distribution unit 34 (S306).

[0113] Here, we will explain in detail the mechanism of the distributed ledger system adopted in the above-mentioned guarantee system. In this embodiment, the guarantee system provides a blockchain interface service, which includes multiple nodes that store at least some or all of the data generated by the AI ​​character management server 3. These nodes aggregate the stored data at predetermined times into blocks, use these blocks to form a blockchain, and share this blockchain among the multiple nodes and store it as a distributed ledger.

[0114] Specifically, as shown in Fig. 7, the AI ​​character provision system 1 according to this embodiment records information for guaranteeing the inexperienced state of the AI ​​character's state parameters in a blockchain via a guarantee system 4. The provision of an AI character according to this embodiment is carried out between two nodes on a P2P (Peer-to-Peer) network 90 (here, between a rights holder and a purchaser), and the transaction information is broadcast and shared among the nodes 90a to 90f within the P2P network 90. ​​As a result, a transaction history database (a so-called blockchain) based on a distributed ledger system is formed on the P2P network 90, and the transaction history of various tokens and the provision of AI characters is stored.

[0115] In this embodiment, the processing by the distributed ledger system involves executing user authentication, approval, and management through the AI ​​character management server 3 when issuing an account specific to each AI character and writing to the limited access rights holder.

[0116] The assurance system 4 generates a public key from the private key based on a digital signature algorithm such as Elliptic Curve Digital Signature Algorithm (ESDSA). Next, the assurance system 4 generates an AI character state assurance public address from the public key by applying a one-way hash function such as SHA-256 or RIPEMD-160 to the public key. For example, the assurance system 4 can generate an AI character state assurance public address by applying SHA-256 to the public key twice. In other words, this AI character state assurance public address is a hash value of the public key used to sign the transaction described above, and is used to identify data related to the AI ​​character.

[0117] Then, data related to the AI ​​character is linked to the AI ​​character's "public address for state assurance" generated in step S303 and recorded in the node. Specifically, information related to the AI ​​character acquired by the assurance system 4, such as the ID of the user who interacted with the AI ​​character while in the general approachable state, information about the user who acquired the limited approach state, and the state parameters at that time, is linked to the AI ​​character's public address for state assurance and made public. This information related to the AI ​​character can be freely viewed by anyone who obtains the public key for the public address for state assurance of the AI ​​character. As a result, anyone can verify whether there has been any fraud or tampering in the history of the power generation method and power generation location from which the electricity originates, or in the transaction history.

[0118] (4) Behavior when checking inexperienced state In this way, the information linked to each AI character's "public address for state assurance" is acquired by a specific user ("user U1" in the above example) as an NFT (Non-Fungible Token) at the time the "limited approach state" is obtained, and only that user will be able to proceed to the next event. On the other hand, any other user who obtains the public key linked to user U1's NFT will be able to view the state linked to the AI ​​character's public address for state assurance PA on the NFT platform, but will not be able to proceed to the next event.

[0119] In this mechanism, an NFT trading application is installed on the smartphone 5 of the acquirer in the limited approach state, and this application manages the state assurance public address PA held by the acquirer. The state assurance public address PA is associated with the acquirer's (user U1) private key, which allows the acquirer to update the NFT from their own state assurance public address PA. This state assurance public address and private key can also be transferred to another person, and each private key allows the acquirer in the limited approach state to freely use the NFT stored in the state assurance public address PA and its transaction history.

[0120] For example, by querying the blockchain, user U1 and other users U2 to U3 can learn that a specific user U1 has acquired the right to a limited approach state. Specifically, as shown in Figure 7, to confirm whether a specific AI character C1 is in an inexperienced state, the smartphone 5 of the user wishing to confirm sends an inexperienced state confirmation request to the AI ​​character management server 3 (S401). In response to this request, the AI ​​character management server 3 executes a setting information query process (S402), and a query request that performs blockchain query processing is sent to the assurance system 4, which acquires information linked to the AI ​​character C1's public address for status assurance PA (S403).

[0121] Next, a state assurance process is executed based on the state parameters acquired in step S403 (S404). This state assurance process generates and distributes evaluation data that evaluates which users have acquired the limited approach state for which AI characters (S405 and S406). Based on this evaluation data, user U1 who has acquired the right to the limited approach state can obtain the service and status to proceed to the next event, while the other users U2 to U3 continue to maintain the broad approachable state, but are set to a state in which the occurrence of the next event is restricted.

[0122] This allows those who have acquired the limited approach status to not only view the service history linked to the status assurance public address PA of this AI character C1, but also receive services unique to the limited approach status, such as proceeding to the next event. Note that this public address may be issued by the AI ​​character management server 3, by software on each trading user's terminal, or by the server of an independent service management institution or financial institution.

[0123] At this time, each block recorded in the NFT stores multiple transactions, a nonce, and the hash value of the previous block. A nonce is a disposable random value used in cryptographic communications, and the node (miner) 90a to 90f that first discovers this value acts as an approver and updates the blockchain by adding the block in which the nonce was discovered to the end of the blockchain.

[0124] As a result, a consistent transaction history is recorded on the blockchain, and by sharing this blockchain among all nodes 90a to 90f participating in the P2P network 90, a consistent transaction history can be shared across the entire P2P network 90. ​​In other words, this blockchain serves as part or all of the NFT transaction history database 41a and key information database 41b in the guarantee system 4 described above. In this embodiment, when providing AI characters based on public key cryptography, NFT transactions are carried out using this mechanism.

[0125] (Actions and Effects) As described above, according to this embodiment, by setting initial values ​​for state parameters that form the AI ​​character's characteristics such as personality, experience, and knowledge, and using the blockchain mechanism to strictly manage the history of changes to these settings, an innocent, inexperienced state uninfluenced by others is guaranteed, allowing the user to enjoy a realistic first experience with the character.

[0126] In addition, the system monitors input and output to the AI ​​character in real time, predicting fluctuations in state parameters to determine whether the character has simply learned various phenomena as knowledge or whether it has gained real-life experience that will influence its subsequent behavior and values, and thereby managing the change history of inexperienced state data.By recording the change history on the blockchain network, the change history can be maintained in an immutable form, preventing data tampering and ensuring the authenticity of the character.

[0127] In particular, by analyzing the recorded change history and calculating the identity with respect to the virgin state data, it is possible to evaluate how much the character has changed from the initial setting. Therefore, even if there is a change history, by calculating the identity between the virgin state and the current state, it is possible to evaluate and guarantee whether an effective virgin state is maintained, thereby preventing unintended changes to the character's characteristics and allowing the user to continue to enjoy a consistent character experience. As a result, according to this embodiment, it is possible to monitor and predict changes in the state of an AI character in real time, and ensure that the AI ​​character is in an inexperienced state.

[0128] The present invention is not limited to the above-described embodiments, and the components can be modified and embodied in practice without departing from the spirit of the invention. Furthermore, various inventions can be created by appropriately combining the components disclosed in the above-described embodiments. For example, some components may be omitted from all the components shown in the embodiments. [Explanation of symbols]

[0129] C1...AI character PA…Public address for status assurance U1...User U2, U3...users 1. AI character provision system 2. Communication network 3. AI character management server 4. Warranty system 5. Smartphone 21…satellite 22...Radio base station 31...Communication interface 32...Location information management department 33...Authentication Department 34…Information Distribution Department 35a...User database 35b…Character Management Database 35c…Event Management Database 36...Character control section 36a...Status parameter management unit 36b…NLP Collaboration Department 36c…AI Character Status Assurance Department 36d…Blockchain Collaboration Department 41a…NFT transaction history database 41b...Key information database 41c...Account database 42...Authentication section 43...Communication interface 44…Rights Management Department 44a…Rights Information Section 44b...Public Address Management Department 44c...Authenticity Verification Department 44d…Data update section 51...Communication interface 52...Input interface 52a...Touch panel 53...Output interface 53a...Display section 54...Application execution unit 55...Memory 90…P2P network 90a~90f...Node 361...Inexperienced state setting section 362... Fluctuation prediction section 363... Rate of change control section 364...Comparison section 365...Inexperienced State Change Section 366...History Recording Section 367…Event Execution Department 368...Condition Assurance Department 540...Synchronization processing unit 541...Interface control unit 541a... Input / output monitoring unit 541b...Experience Level Reference Section 541c…Knowledge database query generation unit 541d…Knowledge database matching section 541e...Event condition determination section 541f...Information request control section 541g...Request filtering section 541h: Response filtering section 541i...Status parameter management section 542...Graphics control unit 543…AI character control unit 544...Location information acquisition unit 544a...Time information acquisition unit 545...Display control unit 545a...GUI control unit 546...Display data generation unit

Claims

1. A system that guarantees a specific state of an AI character, an inexperienced state setting unit that sets initial values ​​of state parameters that form the characteristics of the AI ​​character, including at least one of personality, experience, and knowledge, as inexperienced state data; a fluctuation prediction unit that monitors input and output of the AI ​​character and predicts fluctuations in the state parameters; a history recording unit that periodically records the current values ​​of the state parameters and the predicted fluctuations as change history on a blockchain network; a state assurance unit that analyzes the change history recorded by the history recording unit and calculates identity for the unexperienced state data; An AI character status assurance system comprising:

2. a comparison unit that compares a value of a trigger parameter included in the state parameter with a threshold value that causes a specific event to occur; an event execution unit that executes the specific event in response to a comparison result by the comparison unit; Furthermore, The history recording unit adds to the change history whether or not the specific event has occurred and the number of times the specific event has occurred. The AI ​​character status assurance system according to claim 1.

3. a knowledge database that stores knowledge restriction information for restricting a knowledge range corresponding to current state parameters, and also stores keywords or commands that form prompts for said AI character; a change rate control unit that varies a change rate of the state parameter in response to the occurrence of the specific event; a query generation unit that extracts keywords or commands from a knowledge database based on the current value of the state parameter and the rate of change of the state parameter that has been changed in response to the occurrence of the specific event; an information request control unit that generates an information request requesting an information response from the AI ​​character using the keyword or command extracted by the query generation unit; a filtering unit that compares the response content from the AI ​​character with the knowledge restriction information and corrects the response content that deviates from the knowledge range; 3. The AI ​​character status assurance system according to claim 1 or 2, further comprising:

4. The specific event is Only the first user who executed the next specific event is permitted to execute the next specific event and have a conversation about it, and subsequent users who are subsequent users are restricted from executing the next specific event and having a conversation about it; The knowledge scope of the AI ​​character is updated only through the interaction with the first user. The AI ​​character status assurance system according to claim 2.

5. A program that ensures a specific state of an AI character, comprising: an inexperienced state setting unit that sets initial values ​​of state parameters that form the characteristics of the AI ​​character, including at least one of personality, experience, and knowledge, as inexperienced state data; a fluctuation prediction unit that monitors input and output of the AI ​​character and predicts fluctuations in the state parameters; a history recording unit that periodically records the current values ​​of the state parameters and the predicted fluctuations as change history on a blockchain network; a state assurance unit that analyzes the change history recorded by the history recording unit and calculates identity for the unexperienced state data; An AI character status guarantee program characterized by functioning as:

6. The computer a comparison unit that compares a value of a trigger parameter included in the state parameter with a threshold value that causes a specific event to occur; an event execution unit that executes the occurrence of the specific event in accordance with the comparison result by the comparison unit; It further functions as The history recording unit adds to the change history whether or not the specific event has occurred and the number of times the specific event has occurred.

6. The AI ​​character status guarantee program according to claim 5.

7. The computer a knowledge database that stores knowledge restriction information for restricting a knowledge range corresponding to current state parameters, and also stores keywords or commands that form prompts for said AI character; a change rate control unit that varies a change rate of the state parameter in response to the occurrence of the specific event; a query generation unit that extracts keywords or commands from a knowledge database based on the current value of the state parameter and the rate of change of the state parameter that has been changed in response to the occurrence of the specific event; an information request control unit that generates an information request requesting an information response from the AI ​​character using the keyword or command extracted by the query generation unit; a filtering unit that compares the response content from the AI ​​character with the knowledge restriction information and corrects the response content that deviates from the knowledge range.

7. The AI ​​character status guarantee program according to claim 5 or 6, further functioning as:

8. The specific event is Only the first user who executed the next specific event is permitted to execute the next specific event and have a conversation about it, and subsequent users who are subsequent users are restricted from executing the next specific event and having a conversation about it; The knowledge scope of the AI ​​character is updated only through the interaction with the first user.

7. The AI ​​character status guarantee program according to claim 6.

9. A method for causing a computer to maintain a specific state of an AI character, comprising: an inexperienced state setting step in which an inexperienced state setting unit of the computer sets initial values ​​of state parameters forming characteristics of the AI ​​character, including at least one of personality, experience, and knowledge, as inexperienced state data; a fluctuation prediction step in which a fluctuation prediction unit of the computer monitors input and output of the AI ​​character and predicts fluctuations of the state parameters; a history recording step in which a history recording unit of the computer periodically records the current value of the state parameter and the predicted fluctuation as a change history on a blockchain network; a state assurance step in which a state assurance unit of the computer analyzes the change history recorded by the history recording unit and calculates identity with respect to the unexperienced state data; An AI character state assurance method comprising:

10. a comparison step in which a comparison unit of the computer compares a value of a trigger parameter included in the state parameter with a threshold value for generating a specific event; an event execution step in which an event execution unit of the computer executes the generation of the specific event according to the comparison result by the comparison unit; further comprising In the history recording step, whether or not the specific event has occurred and the number of times the specific event has occurred are added to the change history. The AI ​​character state assurance method according to claim 9.

11. a knowledge storage step in which the computer stores knowledge restriction information for restricting a knowledge range corresponding to current state parameters, and stores keywords or commands forming prompts for the AI ​​character in a knowledge database; a change rate control step in which a change rate control unit of the computer varies a change rate of the state parameter in response to the occurrence of the specific event; a query generation step in which a query generation unit of the computer extracts keywords or commands from a knowledge database based on the current value of the state parameter and the rate of change of the state parameter that has been changed in response to the occurrence of the specific event; an information request control step in which an information request control unit of the computer generates an information request requesting an information response from the AI ​​character using the keyword or command extracted in the query generation step; a filtering unit that compares the response content from the AI ​​character with the knowledge restriction information and corrects the response content that deviates from the knowledge range; The method for ensuring the state of an AI character according to claim 9 or 10, further comprising:

12. The specific event is Only the first user who executed the next specific event is permitted to execute the next specific event and have a conversation about it, and subsequent users who are subsequent users are restricted from executing the next specific event and having a conversation about it; The knowledge scope of the AI ​​character is updated only through the interaction with the first user. The AI ​​character state assurance method according to claim 10.

Images