System

Abstract

A system is provided.SOLUTION: A system comprising: means for generating a profile of an imposter character using a generative AI model; means for generating an interaction scenario between the imposter character and a user; means for presenting the generated interaction scenario to the user and receiving a user response; means for collecting and analyzing the received interaction scenario; and means for improving the generative AI model based on the analyzed scenario.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The technology of the present disclosure relates to a system. [Background technology]

[0002] Patent document 1 discloses a persona chatbot control method performed by at least one processor, the method including the steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to a description of the chatbot character, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance in response to the user utterance. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2022-180282 Summary of the Invention [Problem to be solved by the invention]

[0004] In today's world, where special fraud methods are constantly evolving and becoming increasingly sophisticated, people are at risk of falling prey to fraud. Furthermore, there is a problem that victims suffer serious economic losses and psychological distress when fraudsters illegally obtain personal information and funds. To address these issues, there is a need for a system that provides effective preventative measures and deters fraud victims from becoming victims. [Means for solving the problem]

[0005] The present invention is a system including a means for generating a profile of a fraudster character using a generative AI model, a means for generating a dialogue scenario between the fraudster character and a user, a means for presenting the generated dialogue scenario to the user and receiving the user's response, a means for collecting and analyzing the received dialogue data, and a means for improving the generative AI model based on the analyzed data. Furthermore, the system includes a means for generating a dialogue scenario based on a specific fraud technique to reproduce the fraud technique, and a means for progressing the dialogue scenario with the generated fraudster character in real time, providing a dialogue experience that heightens the user's vigilance, thereby preventing fraud and strengthening countermeasures.

[0006] A "generative AI model" is a system that uses artificial intelligence technology to learn data and automatically generate specific characters and scenarios.

[0007] A "profile" is a collection of information that details the characteristics and attributes of a particular character or individual.

[0008] A "con man character" is a virtual character created using a generative AI model that has the attributes and behavioral patterns of a con man.

[0009] A "dialogue scenario" is a script that pre-determines and generates the flow and content of the dialogue between the user and the virtual character.

[0010] A "user" is an individual or group who uses the system to experience interactive scenarios and learn about fraud methods and countermeasures.

[0011] "Dialogue data" is a collection of information that records the content of a dialogue that took place between a user and a virtual character.

[0012] "Analysis" is the process of using collected interaction data to organize and evaluate information and discover behavioral patterns and trends.

[0013] "Improvement" is the process of making adjustments and updates to improve the performance and functionality of a system or model based on insights gained from analysis.

[0014] "Reproduction" is the act of virtually recreating a specific technique or situation, thereby observing and experiencing its procedures and effects.

[0015] "Increasing vigilance" refers to actions aimed at helping users understand the methods and risks of fraud in advance, and to acquire defensive measures and ways of dealing with actual fraud. [Brief explanation of the drawings]

[0016] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0017] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0018] First, the terms used in the following description will be explained.

[0019] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0020] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0021] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0022] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0023] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0024] [First embodiment]

[0025] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0026] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0027] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0028] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0029] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0030] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0031] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0032] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0033] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0034] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0035] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0036] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0037] This invention relates to a system that uses a generative AI model to generate fraudster characters, and aims to improve users' understanding of fraudulent methods and their vigilance through dialogue between the users and the virtual fraudster characters. This system consists of a server, a terminal, and a user, and each component operates as follows:

[0038] Setting up the con man character

[0039] The server inputs conditions and parameters for generating a con artist character profile into the generative AI model. These conditions include the con artist's age, gender, speech characteristics, behavioral patterns, etc. The generative AI model uses these inputs to generate a specific con artist character profile, which is then stored on the server.

[0040] Interaction generation with virtual characters

[0041] A user accesses the system using a terminal and logs in to begin an impostor experience session. The terminal sends the user's authentication information to the server, which verifies it. If authentication is successful, the server uses a generative AI model to generate an interaction scenario with the impostor character. This scenario is displayed on the user's terminal, and the user begins interacting with it.

[0042] Recreating the fraud method

[0043] The server uses the generative AI model to generate scenarios for recreating specific fraud methods (e.g., phishing scams, technical support scams, etc.). The device presents the generated scenarios to the user in real time and receives the user's responses. This allows the user to interact with the fraudster character and understand the fraud methods and patterns.

[0044] Analysis and improvement of dialogue data

[0045] The server collects and stores dialogue data between users and virtual con artists. This data is fed back to the generative AI model, which analyzes the behavioral patterns of con artists and trends in fraudulent methods. Based on the analysis results, the generative AI model is improved. These improvements make future dialogue scenarios more realistic and effective, improving the accuracy of replicating fraudulent methods.

[0046] Specific examples

[0047] For example, suppose you want to create a male con artist character in his 20s. The server inputs the following conditions into the generative AI model: "male in his 20s, polite speaking, fraudulent technical support." Based on this, the generative AI model generates a specific con artist character. When a user accesses the system from a terminal and starts a con artist experience session, the following dialogue is generated:

[0048] Scammer: "Hello, this is tech support. Are you having trouble with your computer?"

[0049] User: "Yes, something seems to be wrong."

[0050] Scammer: "So, can you grant me remote access?"

[0051] Through these interactions, users can experience the techniques used by technical support scammers and become more vigilant against them. The server collects and analyzes the interaction data and improves the AI ​​model, enabling it to generate more accurate scammer characters and provide more accurate interaction scenarios.

[0052] The processing flow will be explained below.

[0053] Specific explanation of the program's processing steps

[0054] Step 1: Setting up your imposter character

[0055] 1. The server inputs the conditions and parameters for generating a profile of a con man character into the generative AI model.

[0056] Specifically, information such as the scammer's age, gender, language characteristics, and behavioral patterns is set.

[0057] 2. A generative AI model uses these inputs to generate a profile of a specific imposter character.

[0058] The profile includes the scammer's background, the fraudulent methods they use, and how they lead you to the scam.

[0059] 3. The server stores the generated profile in a database.

[0060] Step 2: Authenticating the user and starting an interactive session

[0061] 1. The user accesses the system from a terminal and logs in.

[0062] The user enters authentication information (e.g., username and password) through a dedicated interface.

[0063] 2. The terminal sends a login request to the server.

[0064] 3. The server checks the database and, if authentication is successful, allows the interactive session to begin.

[0065] If the authentication fails, an error message is returned to the terminal.

[0066] Step 3: Creating interactions with virtual characters

[0067] 1. The user operates the device and sends a request to start an "impostor experience session."

[0068] The user clicks the "Start Trial Session" button within the interface.

[0069] 2. The server uses the generative AI model to generate dialogue scenarios with the impostor character.

[0070] The generative AI model creates initial dialogue scenarios and response patterns based on the profile of the conman character.

[0071] 3. The terminal displays the dialogue scenario to the user in real time and receives the user's responses.

[0072] Responses from the con man character are generated and displayed sequentially in response to user input.

[0073] Step 4: Recreate the scam

[0074] 1. The server uses a generative AI model to generate scenarios to recreate specific fraud schemes.

[0075] The server tells the AI ​​model what type of fraud scheme it wants to replicate, such as a "phishing scam" or a "tech support scam."

[0076] 2. The generative AI model generates dialogue scenarios based on the specified fraudulent tactics.

[0077] The scenario includes specific steps on how the scammer will deceive the target.

[0078] 3. The terminal displays the generated fraud scenario to the user and conducts a dialogue in real time.

[0079] By following the scenario, users can simulate the process of fraudulent schemes.

[0080] Step 5: Collect and analyze interaction data

[0081] 1. The server collects and stores in a database the interaction data between the user and the virtual impostor character.

[0082] All messages and responses that occur during each interactive session are collected as logs.

[0083] 2. The server applies a data analysis algorithm to analyze the collected interaction data.

[0084] Identify trends in fraudster behavior and tactics.

[0085] Step 6: Improving the generative AI model

[0086] 1. Based on the analysis results, the server provides improvement feedback to update the algorithm of the generative AI model.

[0087] The algorithm of the generative AI model is updated based on new patterns and techniques, and these are reflected in the next dialogue generation.

[0088] 2. The server deploys the system to a new version, providing users with the latest fraud techniques and countermeasures.

[0089] The system is automatically updated to reflect the latest fraud techniques and countermeasures.

[0090] In this way, the system not only allows users to experience real interactions with scammers, thereby raising their vigilance, but also allows users to analyze the generated data to understand scammers' behavioral patterns and new techniques, further improving the generative AI model.

[0091] Example 1

[0092] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0093] There is a lack of systems to raise awareness and vigilance against real-world fraud methods. While the risk of being victimized increases when personal information is carelessly disclosed, there are limited ways to learn effective prevention measures without experiencing actual fraud methods. In addition, there is a need for technology that can quickly respond to fraud methods that are constantly evolving.

[0094] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0095] In this invention, the server includes means for generating a profile of a fraudster character using a generative AI model, means for generating a dialogue scenario between the fraudster character and a user, means for presenting the generated dialogue scenario and receiving the user's responses, means for collecting and analyzing the received dialogue data, means for improving the generative AI model based on the analyzed data, means for receiving and authenticating authentication information, and means for updating the dialogue scenario based on the user's responses. This allows users to safely and effectively experience and learn realistic fraud techniques. Furthermore, the continuous improvement process using the generative AI model provides practical learning that always keeps up with the latest fraud techniques.

[0096] A "generative AI model" is an artificial intelligence technology that learns from a variety of data and generates new characters and dialogue scenarios.

[0097] A "profile" is information that details the characteristics and attributes of a particular character.

[0098] A "scammer character" is a virtual character that imitates fraudulent methods and reproduces the fraudulent tactics through dialogue with users.

[0099] A "dialogue scenario" is a script that comprises a series of dialogues that take place between the con man character and the user.

[0100] "User" means any person or entity that uses the System.

[0101] "Authentication information" refers to information used to verify a user's identity, and includes, for example, a username and password.

[0102] "Received dialogue data" is information regarding the dialogue that took place between the user and the impostor character.

[0103] "Collection and analysis" is the process of gathering interaction data and examining its content and trends.

[0104] "Improving a generative AI model" is the process of improving the performance and accuracy of an AI model based on collected and analyzed data.

[0105] "Updating" means to generate a new dialogue scenario based on the user's response and continue the dialogue scenario.

[0106] "Authentication" is the process of verifying that a user has valid access rights.

[0107] The system of the present invention comprises a server, a terminal, and a user. A specific embodiment will be described below.

[0108] Setting up the con man character

[0109] The server first generates a profile for the con man character using a generative AI model. Parameters such as age, gender, speech characteristics, and behavioral patterns are used to generate the profile. Specifically, the server inputs the following prompt sentence into the generative AI model:

[0110] "Generate a male in his 20s, polite-spoken, fraudulent technical support scammer character."

[0111] The generative AI model generates a profile of the imposter character based on this prompt sentence and saves the profile on the server.

[0112] User Login and Authentication

[0113] A user accesses the system using a terminal and enters authentication information (e.g., username and password) on the login screen. The terminal then sends this authentication information to the server.

[0114] The server checks the received authentication information against the database, and if authentication is successful, allows the user access. If invalid authentication information is entered, an error message is returned to the terminal.

[0115] Start of the Impostor Experience Session

[0116] After the user is successfully authenticated, they send a request to the server to start an impostor experience session. The server then sends a request to the generative AI model to generate a dialogue scenario with the impostor character.

[0117] The generative AI model generates a dialogue scenario for the con man character and returns the scenario to the server, which then sends the generated scenario to the device.

[0118] Recreating the fraud method

[0119] The terminal presents the dialogue scenario received from the server to the user in real time, and the user inputs a response according to the dialogue scenario, which is then sent to the server by the terminal.

[0120] The server receives the user's response and requests the generative AI model to continue the dialogue based on that response. The generative AI model generates a new dialogue and returns it to the server. The server then sends this new dialogue to the device. This process is repeated, and the user experiences a dialogue with the virtual impostor character.

[0121] Analysis and improvement of dialogue data

[0122] The server collects and stores all interaction data between the user and the virtual impostor character, which is stored in a database for analysis.

[0123] The server uses the collected dialogue data to provide feedback to the generative AI model, improving the behavior of the conman character and the algorithm for generating dialogue scenarios, making future dialogue scenarios more realistic and effective.

[0124] Specific examples

[0125] For example, to create a male con man character in his 20s, the server inputs conditions such as "male in his 20s, polite speech, fraudulent technical support" into the generative AI model. The generative AI model generates a specific con man character based on these conditions. When a user accesses the system from a terminal and starts a con man experience session, the following dialogue is generated:

[0126] Scammer: "Hello, this is tech support. Are you having trouble with your computer?"

[0127] User: "Yes, something seems to be wrong."

[0128] Scammer: "So, can you grant me remote access?"

[0129] Through these interactions, users can experience tech support scams and become more vigilant against them. The server collects and analyzes the interaction data and improves the generative AI model, enabling it to generate more accurate scammer characters and provide more accurate interaction scenarios.

[0130] Summary of prompt sentence examples

[0131] "Generate a male in his 20s, polite-spoken, fraudulent technical support scammer character."

[0132] "Generate a phishing scenario"

[0133] "Continue the dialogue of the imposter character based on the user's response."

[0134] This will enable users to learn specific countermeasures against various fraudulent methods and create a system that can prevent fraud damage before it occurs.

[0135] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0136] Step 1:

[0137] The server inputs a prompt sentence for generating a profile to the generative AI model.

[0138] Input: Character attribute information (e.g., age, gender, speech characteristics, behavior patterns)

[0139] Data processing: The generative AI model generates a character profile based on the input attribute information.

[0140] Output: Generated character profile

[0141] Specific operation: For example, a prompt such as "Male in his 20s, polite speaking, fraudulent technical support" is sent to the generation AI model, and a profile of the fraudster character is generated and stored on the server.

[0142] Step 2:

[0143] A user accesses the system using a terminal and enters authentication information (e.g., username and password) on a login screen.

[0144] Input: User authentication information (username, password)

[0145] Data processing: The terminal sends the entered authentication information to the server.

[0146] Output: Sending authentication information to the server

[0147] Specific operation: The user enters the username and password on the login screen and clicks the "Login" button.

[0148] Step 3:

[0149] The server checks the received authentication information against a database and allows the user access if authentication is successful.

[0150] Input: User authentication information sent from the device

[0151] Data processing: Check against the database to determine whether authentication was successful.

[0152] Output: Permissions or error messages

[0153] Specific operation: The server checks the authentication information stored in the database, and if it matches, notifies the terminal that authentication was successful.

[0154] Step 4:

[0155] After the user is successfully authenticated, the terminal sends a request to the server to initiate an impostor experience session.

[0156] Input: Session initiation request

[0157] Data processing: The server receives the request and begins processing.

[0158] Output: Session start confirmation message

[0159] Specific action: The user clicks the "Start Impostor Experience Session" button and submits the request.

[0160] Step 5:

[0161] The server sends a request to the generative AI model to generate an interaction scenario with the impostor character.

[0162] Input: Impostor Experience Session Start Request

[0163] Data processing: The generative AI model generates dialogue scenarios with the con man character.

[0164] Output: Generated dialogue scenario

[0165] Specific operation: The generative AI model generates a "tech support scam" scenario and returns it to the server.

[0166] Step 6:

[0167] The terminal presents the dialogue scenario received from the server to the user in real time.

[0168] Input: Interaction scenario sent from the server

[0169] Data processing: The scenario received by the terminal is displayed on the screen.

[0170] Output: Display of the interaction scenario to the user

[0171] What it does: The device displays the scammer character's line, "Hello, this is tech support. Are you having trouble with your computer?"

[0172] Step 7:

[0173] The user inputs a response according to the dialogue scenario, and the response is transmitted by the terminal to the server.

[0174] Input: User response (e.g. "Yes, something seems to be wrong.")

[0175] Data processing: The terminal sends the user's response to the server.

[0176] Output: Sending the user response to the server

[0177] Specific behavior: The user types a response and clicks the send button.

[0178] Step 8:

[0179] The server receives the user's response and, based on that, requests the generative AI model to continue the dialogue.

[0180] Input: User response

[0181] Data processing: Ask the generative AI model to continue the dialogue and generate a new scenario.

[0182] Output: Generated new dialogue scenario

[0183] Specific behavior: The generative AI model generates a new dialogue scenario and generates the reply, "So, may I grant you remote access?"

[0184] Step 9:

[0185] The server transmits the generated new dialogue scenario to the terminal.

[0186] Input: Generated new dialogue scenario

[0187] Data processing: Send the dialogue scenario to the terminal.

[0188] Output: Sending the dialogue scenario to the terminal

[0189] Specific operation: The server sends the newly generated dialogue scenario to the terminal, which then displays it to the user.

[0190] Step 10:

[0191] The server collects, stores, and analyzes all interaction data between the user and the virtual impostor character.

[0192] Input: Interaction data between the user and the impostor character

[0193] Data processing: Collection, storage, and analysis of interaction data

[0194] Output: Analysis results

[0195] What it does: The server stores the collected interaction data in a database and uses data analysis tools to extract patterns.

[0196] Step 11:

[0197] The server provides feedback to the generative AI model based on the analysis results, improving the AI ​​model.

[0198] Input: Analysis results

[0199] Data processing: Updating generative AI models

[0200] Output: An improved generative AI model

[0201] How it works: The generative AI model is refined based on newly collected data and analysis results, improving the quality of future dialogue scenarios.

[0202] (Application example 1)

[0203] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0204] Fraud methods continue to become more sophisticated, increasing the risk of many people becoming victims of fraud. Fraud methods via the Internet and telephone are particularly complex, making it difficult for many people to experience the risks in a realistic manner and raise their vigilance. Therefore, there is a need for a system that allows users to experience actual fraud methods and raise their vigilance.

[0205] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0206] In this invention, the server includes means for generating a profile of a fraudster character using a generative AI model, means for generating a dialogue scenario between the fraudster character and a user, and means for providing the generated dialogue scenario in a smartphone application, thereby enabling users to experience fraudulent tactics in a realistic manner through their smartphones and increase their vigilance.

[0207] A "generative AI model" is an artificial intelligence model that extracts specific patterns and information from data and generates natural language.

[0208] A "scammer character" is a profile of a virtual scammer created by a generative AI model, including age, gender, language, and behavioral patterns.

[0209] A "dialogue scenario" is a script that shows the flow of a pre-set conversation that takes place between a con man character and a user.

[0210] A "smartphone application" is a part of a program that runs on a smartphone and provides an interface for a user to interact with an impostor character.

[0211] "User response" refers to a reply or action that a user inputs or selects in a dialogue scenario.

[0212] "Dialogue data" is a record of the dialogue that took place between the user and the impostor character.

[0213] "Improving the generative AI model" refers to the process of analyzing collected interaction data and using the results to further improve the generative AI model.

[0214] A "fraudulent technique" is a specific method or means used by a fraudster to deceive a victim.

[0215] "Interactive experience to increase vigilance" refers to an educational dialogue that allows users to virtually experience fraudulent methods in order to increase their vigilance against real-life scams.

[0216] This invention relates to a system that uses a generative AI model to generate fraudster characters, allowing users to virtually experience fraud and raise their vigilance. The system consists of a server, a terminal (such as a smartphone), and a user.

[0217] Server Processing

[0218] The server uses a generative AI model to generate a profile of the fraudster character. To generate the profile, the user inputs conditions such as the fraudster's age, gender, speech characteristics, and behavioral patterns. This generates a specific fraudster character and stores it on the server.

[0219] The server then uses the generative AI model to generate a dialogue scenario for the user. This dialogue scenario is generated based on a specific fraudulent scheme (e.g., technical support fraud or phishing scam). The generated scenario is saved on the server and then sent to the user's device (smartphone).

[0220] Terminal (smartphone) processing

[0221] A user accesses the system using a smartphone and logs in. After logging in, the device sends the user's authentication information to the server, and authentication is performed. If authentication is successful, the server sends a dialogue scenario with the impostor character to the device and presents it to the user.

[0222] When the user initiates a dialogue, the device receives the user's response and sends it to the server. As the dialogue between the user and the impostor character progresses, the device continues to update the scenario in real time.

[0223] User Experience

[0224] The user experiences a virtual fraud through a smartphone application. For example, if the user selects the "tech support fraud" scenario, the following dialogue is generated:

[0225] Scammer Character: "Hello, tech support. Having trouble with your computer?"

[0226] User: "Yes, something seems to be wrong."

[0227] Impostor Character: "So, can you grant me remote access?"

[0228] Through such interactions, users can develop a sense of vigilance against real scams.

[0229] Data collection and analysis

[0230] Once the conversation ends, the server collects and stores the conversation data with the user. This data is fed back to the generative AI model, which analyzes the behavioral patterns of fraudsters and trends in fraudulent methods. Based on the analysis results, the generative AI model is improved. These improvements make future conversation scenarios more realistic and effective, improving the accuracy of replicating fraudulent methods.

[0231] Prompt Sentence Examples

[0232] Here's an example of a specific prompt for a generative AI model to generate a profile for a con man character:

[0233] "Male, 25 years old, polite speaking, fraudulent tech support scam. Generate a scenario to convince a user to grant remote access."

[0234] The present invention provides an effective means for users to experience virtual fraud through a smartphone application, thereby increasing their vigilance against real-life fraud.

[0235] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0236] Step 1:

[0237] The server inputs a prompt to the generative AI model to generate a profile of a con man character. This prompt includes the con man's age, gender, language, and behavioral patterns. Based on these conditions, the generative AI model generates a profile of a specific con man character and saves the profile on the server. The input data is the prompt, and the output data is the con man character's profile.

[0238] Step 2:

[0239] The server uses the generative AI model to generate a dialogue scenario between the conman character and the user. Specifically, the server inputs data on the conman character's profile and fraudulent methods into the generative AI model, and outputs a dialogue scenario. The generated dialogue scenario is saved on the server. The input data is the conman character's profile and fraudulent methods, and the output data is the dialogue scenario.

[0240] Step 3:

[0241] A user accesses the system using a smartphone terminal and logs in. The terminal sends the user's authentication information (username and password) to the server, which then authenticates it. If authentication is successful, the server sends the generated dialogue scenario to the terminal. The input data is the user's authentication information, and the output data is the authentication result and the dialogue scenario.

[0242] Step 4:

[0243] When the user selects to start the scenario, the smartphone device begins a dialogue with the imposter character. The user inputs a response to the imposter character's prompt, which is then sent from the device to the server. The server receives the user's response and updates the dialogue scenario as necessary. The input data is the user's response, and the output data is the updated dialogue scenario.

[0244] Step 5:

[0245] Once the interaction is complete, the server collects and stores the interaction data with the user. This interaction data includes the user's responses and the statements made by the imposter character. The server analyzes this data and feeds it back into the generative AI model. The input data is the interaction data, and the output data is the analysis results.

[0246] Step 6:

[0247] The server improves the generative AI model based on the analysis results. This makes the generation of future dialogue scenarios more realistic and effective, improving the accuracy of reproducing fraudulent methods. The input data is the analysis results, and the output data is the improved generative AI model.

[0248] Through the above processing steps, users can virtually experience fraud through a smartphone application, increasing their vigilance against real-life fraud.

[0249] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0250] This invention relates to a system that generates fraudster characters by combining a generative AI model and an emotion engine, and aims to improve users' understanding of fraudulent methods and their vigilance through dialogue between the user and the virtual fraudster character. This system consists of a server, a terminal, a user, and an emotion engine, and each component operates as follows:

[0251] Setting up the con man character

[0252] The server inputs conditions and parameters for generating a con artist character profile into the generative AI model. These conditions include the con artist's age, gender, speech characteristics, behavioral patterns, etc. The generative AI model uses these inputs to generate a specific con artist character profile, which is then stored on the server.

[0253] Authenticating the user and starting an interactive session

[0254] A user accesses the system using a terminal and logs in to begin an impostor experience session. The terminal sends the user's authentication information to the server, which verifies it. If authentication is successful, the server uses a generative AI model to generate an interaction scenario with the impostor character. This scenario is displayed on the user's terminal, and the user begins interacting with it.

[0255] Incorporating an emotion engine

[0256] The emotion engine analyzes video and audio data acquired from the user's camera and microphone to recognize the user's real-time emotional state. This emotional data is sent to the server and reflected in the responses of the con man character and the progression of the scenario.

[0257] Fraud reenactment and emotional feedback

[0258] The server uses the generative AI model to generate scenarios for recreating specific fraud schemes (e.g., phishing scams, technical support scams, etc.). The device presents the generated scenarios to the user in real time, and the emotion engine analyzes the user's emotional state. This allows the fraudster character to respond appropriately based on the user's emotional data. For example, if the user is feeling anxious, the fraudster character will use words that will further reassure them.

[0259] Collection and analysis of dialogue data

[0260] The server collects dialogue and emotional data between the user and the virtual con man character and stores it in a database. This data is fed back to the generative AI model and emotion engine, which analyzes trends in the con man's behavioral patterns and emotional responses. Based on the analysis results, both models are improved. These improvements make future dialogue scenarios more realistic and effective, improving the accuracy of reproducing fraudulent methods.

[0261] Specific examples

[0262] For example, let's create a male scammer character in his 20s on the theme of technical support fraud. The server inputs the following conditions into the generative AI model: "male in his 20s, polite speech, fraudulent technical support." Based on this, the generative AI model generates a specific scammer character. When a user accesses the system from a terminal and starts a scammer experience session, the following dialogue is generated:

[0263] Scammer: "Hello, this is tech support. Are you having trouble with your computer?"

[0264] User: "Yes, something seems to be wrong."

[0265] Scammer: "So, can you grant me remote access?"

[0266] During this dialogue, the emotion engine analyzes the user's facial expressions and voice to recognize their emotions in real time. If the emotion engine determines that the user is feeling anxious, the con man character will respond with a reassuring message, saying, "Remote operation is very safe and we can resolve the issue immediately." The server collects dialogue data and emotion data and feeds it back to the generative AI model and emotion engine to improve the accuracy of the system.

[0267] In this way, the system not only allows users to experience real interactions with scammers, thereby raising their vigilance, but also analyzes the generated data to understand scammers' behavioral patterns and new techniques, further improving the generative AI model and emotion engine.

[0268] The processing flow will be explained below.

[0269] Specific explanation of the program's processing steps

[0270] Step 1: Setting up your imposter character

[0271] 1. The server inputs the conditions and parameters for generating a profile of a con man character into the generative AI model.

[0272] The criteria include age, gender, language characteristics, and behavioral patterns.

[0273] 2. The generative AI model uses these conditions to generate a profile of a specific imposter character.

[0274] The profile includes the scammer's background, the fraudulent methods they use, and their luring techniques.

[0275] 3. The server stores the generated profile in a database.

[0276] Step 2: Authenticating the user and starting an interactive session

[0277] 1. A user accesses the system using a terminal and logs in.

[0278] The user enters authentication information (user name, password) in a dedicated interface.

[0279] 2. The terminal sends a login request to the server.

[0280] 3. The server checks the database and, if authentication is successful, allows the interactive session to begin.

[0281] If the authentication fails, an error message is returned to the terminal.

[0282] Step 3: Creating interactions with virtual characters

[0283] 1. The user operates the device and sends a request to the server to start a "scammer experience session."

[0284] The user clicks the "Start Trial Session" button within the interface.

[0285] 2. The server uses the generative AI model to generate dialogue scenarios with the impostor character.

[0286] The generative AI model creates initial dialogue scenarios and response patterns based on the profile of the conman character.

[0287] 3. The terminal displays the dialogue scenario to the user in real time and receives the user's responses.

[0288] Responses from the con man character are generated and displayed sequentially in response to user input.

[0289] Step 4: Replaying the scam and emotional feedback

[0290] 1. The server uses a generative AI model to generate scenarios to recreate specific fraud schemes.

[0291] The server instructs the AI ​​model on the type of fraudulent scheme it wants to replicate, such as "phishing scams" or "tech support scams."

[0292] 2. The generative AI model generates dialogue scenarios based on the specified fraudulent tactics.

[0293] The scenarios include specific steps on how the scammer will deceive the target.

[0294] 3. The emotion engine analyzes video and audio data obtained from the user's camera and microphone to recognize the user's emotional state.

[0295] Emotions are classified as, for example, "relief," "anxiety," and "doubt."

[0296] 4. The device displays the generated fraud scenario to the user, and the emotion engine analyzes the user's emotional state and transmits it to the server in real time.

[0297] The server then adjusts the imposter character's responses based on this: if the user is feeling anxious, the imposter character will respond in a reassuring way.

[0298] Step 5: Collect and analyze interaction data

[0299] 1. The server collects and stores in a database the conversation data and emotion data exchanged between the user and the impostor character.

[0300] All messages and emotional responses from each interaction session are logged.

[0301] 2. The server applies data analysis algorithms to analyze the collected interaction data and emotion data.

[0302] Identify trends in fraudster behavior patterns and user emotional responses.

[0303] Step 6: Improving the generative AI model and emotion engine

[0304] 1. Based on the analysis results, the server provides feedback to update the generative AI model algorithm and the emotion engine analysis.

[0305] The algorithm of the generative AI model is updated based on new patterns and techniques, and is reflected in the next dialogue generation. The emotion engine is also updated to improve the accuracy of the user's emotional response.

[0306] 2. The server deploys the system to a new version and provides users with an interactive experience that helps them understand and be aware of the latest fraud techniques.

[0307] The system is automatically updated to reflect the latest fraud techniques and countermeasures.

[0308] In this way, the system not only allows users to experience real-life interactions with scammers, thereby raising their vigilance, but also analyzes the generated data to understand scammers' behavioral patterns and new techniques, further improving the generative AI model and emotion engine.

[0309] Example 2

[0310] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0311] With the advancement of modern advanced communication and information technologies, fraud methods are becoming more sophisticated. In particular, fraud methods that skillfully exploit users' psychological state can cause serious damage. However, many users are not sufficiently vigilant against fraudulent methods and are easily deceived. Therefore, effective measures are needed to help users better understand fraudulent methods and increase their vigilance.

[0312] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0313] In this invention, the server includes means for generating a profile of a fraudster character using a generative AI model, means for generating a dialogue scenario between the fraudster character and the user, means for presenting the generated dialogue scenario to the user and receiving the user's response, means for acquiring real-time emotional data of the user using an emotion engine that recognizes the user's emotional state, means for collecting and analyzing the received dialogue data and emotional data, and means for improving the generative AI model based on the analyzed data. This allows the user to learn fraudulent tactics through dialogue with the fraudster character and to increase their vigilance against real-world fraud by receiving feedback on their emotional state in real time.

[0314] A "generative AI model" is a model that uses artificial intelligence technology to generate characters and dialogue scenarios tailored to specific purposes.

[0315] A "scammer character" is an artificially generated character that simulates fraudulent methods, and allows the user to learn fraudulent techniques and methods through dialogue with the user.

[0316] A "profile" is a set of information that includes characteristics such as the conman character's age, gender, speech patterns, and behavioral patterns.

[0317] A "dialogue scenario" specifically defines the flow and content of a series of dialogues that take place between a con man character and a user.

[0318] The "emotion engine" is a system that analyzes the user's facial expressions and voice through devices such as cameras and microphones, and recognizes their emotional state at that time in real time.

[0319] "Real-time" refers to reactions and analysis that occur immediately on the spot while the user is using the system.

[0320] "Emotion data" is information about the emotional state obtained from the user's facial expressions and voice analyzed by the emotion engine.

[0321] "Dialogue data" is information about the content of the conversation between the user and the impostor character, including the flow of the dialogue and specific wording.

[0322] "Analysis" refers to the process of using collected interaction and emotion data to evaluate patterns and trends in order to improve the performance of generative AI models and emotion engines.

[0323] This invention relates to a system that combines a generative AI model and an emotion engine to generate a fraudster character, and aims to improve understanding of fraudulent methods and vigilance through dialogue between the user and the virtual fraudster character. This system consists of a server, a terminal, a user, and an emotion engine, and each component operates as follows:

[0324] Setting up the con man character

[0325] The server inputs the conditions and parameters for generating a scammer character profile into the generative AI model. These conditions include the scammer's age, gender, speech characteristics, and behavioral patterns. For example, a prompt might read, "Please generate a character who is a man in his 20s who speaks politely and commits technical support scams." Based on these inputs, the generative AI model generates a profile for a specific scammer character, and the profile is stored on the server.

[0326] Authenticating the user and starting an interactive session

[0327] A user accesses the system using a terminal and logs in to begin an imposter experience session. The terminal sends the user's authentication information to the server, which verifies it. If authentication is successful, the server uses the generative AI model to generate an interaction scenario with the imposter character. For example, this can be achieved by sending the generative AI model a prompt such as "Please generate a tech support fraud scenario." This scenario is then displayed on the user's terminal, and the user begins interacting with it.

[0328] Incorporating an emotion engine

[0329] The emotion engine analyzes video and audio data acquired from the user's camera and microphone to recognize the user's real-time emotional state. This emotional data is sent to the server and reflected in the responses of the con man character and the progression of the scenario.

[0330] Fraud reenactment and emotional feedback

[0331] The server uses the generative AI model to generate scenarios for recreating specific fraud schemes (e.g., phishing scams, technical support scams, etc.). The device presents the generated scenarios to the user in real time, and the emotion engine analyzes the user's emotional state. This allows the fraudster character to respond appropriately based on the user's emotional data. For example, if the user is feeling anxious, the fraudster character will use words that will further reassure them.

[0332] Collection and analysis of dialogue data

[0333] The server collects dialogue and emotional data between the user and the virtual con man character and stores it in a database. This data is fed back to the generative AI model and emotion engine, which analyzes trends in the con man's behavioral patterns and emotional responses. Based on the analysis results, both models are improved. These improvements make future dialogue scenarios more realistic and effective, improving the accuracy of reproducing fraudulent methods.

[0334] Specific examples

[0335] For example, if you create a male scammer character in his 20s on the theme of technical support fraud, the server inputs the following conditions into the generative AI model: "male in his 20s, politely speaking, technical support fraud." Based on this, the generative AI model generates a specific scammer character.

[0336] When a user accesses the system from a terminal and starts an imposter experience session, the following dialogue is generated:

[0337] Scammer: "Hello, this is tech support. Are you having trouble with your computer?"

[0338] User: "Yes, something seems to be wrong."

[0339] Scammer: "So, can you grant me remote access?"

[0340] During this dialogue, the emotion engine analyzes the user's facial expressions and voice to recognize their emotions in real time. For example, if the emotion engine determines that the user is feeling anxious, the con artist character will respond with a reassuring "Remote operation is very safe, and we can resolve the issue immediately." The server collects dialogue and emotion data and feeds it back to the generative AI model and emotion engine to improve the system's accuracy. In this way, users not only experience realistic dialogue with con artists and become more vigilant, but also analyze the generated data to understand the con artists' behavioral patterns and new techniques, allowing them to make future dialogue scenarios more realistic and effective.

[0341] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0342] Step 1:

[0343] Setting up the con man character

[0344] The server inputs conditions and parameters for generating a profile of a con man character into the generative AI model. The input conditions include the con man's age, gender, speech characteristics, and behavioral patterns. The generative AI model then generates a profile of the con man character based on those conditions. The generated profile is stored in the server's database.

[0345] Specific operation: The server sends the prompt "Generate a character who is a man in his 20s and who will commit technical support scams in a polite tone" to the generation AI model and saves the profile of the resulting scammer character.

[0346] Input: Criteria for the conman character (age, gender, speech characteristics, behavioral patterns)

[0347] Output: Impostor character profile

[0348] Step 2:

[0349] Authenticating Users

[0350] A user accesses the system using a terminal and enters authentication information on the login screen. The terminal sends the authentication information to the server. The server compares the received authentication information with a database and determines whether the authentication is successful or not. If the authentication is successful, the server sends a session start notification to the user.

[0351] How it works: The user enters their username and password on the login screen, and the device sends this to the server, which then checks the information against the database and returns the results.

[0352] Input: User authentication information (username, password)

[0353] Output: Authentication result (success or failure)

[0354] Step 3:

[0355] Start of the Impostor Experience Session

[0356] The server generates a dialogue scenario with the impostor character using the generative AI model. The generated dialogue scenario is sent to the user's device and displayed. The user then begins a dialogue with the impostor character based on the scenario.

[0357] Specific operation: The server sends a prompt to the generative AI model saying, "Please generate a technical support fraud scenario," and presents the resulting scenario to the user's device.

[0358] Input: Prompt sentence (instruction to generate a dialogue scenario with the impostor character)

[0359] Output: Generated dialogue scenario

[0360] Step 4:

[0361] Incorporating an emotion engine

[0362] The device captures video and audio data from the user's camera and microphone. The emotion engine analyzes this data and recognizes the user's emotional state in real time. This emotional data is then sent to the server.

[0363] Specific operation: The device sends video and audio data to the emotion engine, which analyzes it to generate real-time emotion data and send it to the server.

[0364] Input: Video data, audio data

[0365] Output: User emotion data

[0366] Step 5:

[0367] Fraud reenactment and emotional feedback

[0368] The server uses the generative AI model to generate scenarios for recreating specific fraud schemes, which are then presented to the user, and the responses of the fraudster character are adjusted based on data from the emotion engine.

[0369] Specific operation: The server sends a prompt to the generative AI model to "generate a scenario that reproduces a fraudulent scheme," and presents the generated scenario to the user in real time. Based on the emotional data, the fraudster character's responses are appropriately adjusted.

[0370] Input: prompt sentence (instructions for reproducing the fraud scheme), user emotion data

[0371] Output: Adjusted impostor character response

[0372] Step 6:

[0373] Collection and analysis of dialogue data

[0374] The server collects and stores the conversation and emotion data between the user and the impostor character in a database. The collected data is fed back to the generative AI model and emotion engine, which then improves both models.

[0375] How it works: The server collects dialogue and emotion data in real time, stores it in a database, analyzes it, and uses it as feedback to improve the performance of the generative AI model and emotion engine.

[0376] Input: Dialogue data, emotion data

[0377] Output: Improved generative AI model, emotion engine

[0378] (Application example 2)

[0379] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0380] In modern society, fraudsters' tactics are becoming increasingly sophisticated, increasing the risk of ordinary users being scammed. Therefore, there is a need for effective training methods to help users understand actual fraudulent methods and increase their vigilance. However, current technology makes it difficult for users to interactively experience realistic fraud scenarios and receive feedback based on their emotional state while increasing their vigilance. To solve this problem, the development of an innovative system that combines a generative AI model and an emotion engine is desired.

[0381] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0382] In this invention, the server includes: means for generating a profile of a fraudster character using a generative AI model; means for generating a dialogue scenario between the fraudster character and a user; means for presenting the generated dialogue scenario to the user and receiving the user's response; means for collecting and analyzing the received dialogue data and the user's emotional data; means for improving the generative AI model and the emotion engine based on the analyzed data; and means including an emotion engine for adjusting the dialogue scenario based on the user's emotional data. This allows the user to experience fraudulent techniques in real time and receive appropriate feedback according to their emotional state, effectively increasing their vigilance against fraud.

[0383] A "generative AI model" is a form of generative artificial intelligence (AI) that is an algorithm that learns from large amounts of data and generates new profiles or text based on specific conditions and parameters.

[0384] A "scammer character" is a virtual scammer profile constructed using a generative AI model, and is a person with specific attributes such as age, gender, language, and behavioral patterns.

[0385] A "dialogue scenario" refers to the sequence or story of a dialogue that takes place between a con man character constructed by a generative AI model and the user.

[0386] An "emotion engine" is a general term for algorithms and software that analyze a user's facial expressions, voice, actions, etc. in real time and recognize their emotional state.

[0387] "Emotional Data" refers to information about a user's emotional state obtained and analyzed by the emotion engine.

[0388] "Dialogue data" refers to the record of the dialogue between the user and the virtual con man character and the text data thereof.

[0389] A "profile" refers to detailed information about a person generated based on specific conditions or parameters, including, for example, age, gender, speech, and behavioral patterns.

[0390] "Real-time" refers to processing and responses occurring almost immediately, without delay, and indicates a state in which dialogue with the user and emotional analysis are carried out immediately.

[0391] "Feedback" refers to the return of information that allows a generative AI model or emotion engine to change its response or improve the system based on the user's reaction or emotional state.

[0392] The present invention is a system that combines a generative AI model and an emotion engine to help users understand fraudulent tactics and raise their vigilance through dialogue with virtual fraudster characters. The system has the following specific configuration and operation procedures:

[0393] Server Roles

[0394] The server uses a generative AI model to generate a profile for the scammer character by inputting prompt statements containing criteria such as "male in his 20s, polite speech, technical support scam" into the model.

[0395] example:

[0396] text

[0397] Create a scammer character with the following attributes: 20-year-old male with polite speech, specializing in technical support scams.

[0398] Furthermore, the server uses the generative AI model to generate dialogue scenarios between the scammer character and the user, based on scamming techniques (e.g., phishing scams and tech support scams).

[0399] Device Role

[0400] The terminal used by the user is a device such as a smartphone or a head-mounted display. The terminal presents the dialogue scenario sent from the server to the user and receives the user's response.

[0401] The device also inputs video and audio data acquired from the user's camera and microphone into an emotion engine to recognize the user's real-time emotional state. The emotion engine uses software such as Affectiva. The emotion data is sent to the server and reflected in the responses of the imposter character and the progression of the scenario. For example, if the user shows signs of anxiety, the imposter character will respond reassuringly, saying, "Remote operation is very safe and we can solve the problem immediately."

[0402] User Roles

[0403] Users access the system via a terminal, log in, and begin a fraudster experience session. During the session, users interact with a virtual fraudster character and respond according to the progression of the dialogue scenario. Through this interaction, users can experience realistic fraudulent tactics and become more vigilant.

[0404] The server collects and analyzes user response and emotion data. The analyzed data is fed back to the generative AI model and emotion engine. This further improves the generative AI model and emotion engine, making future dialogue scenarios more realistic and effective.

[0405] These configurations and operational procedures allow users to experience fraudulent techniques in real time while receiving appropriate feedback based on their emotional state, which can increase their vigilance against fraud and help prevent them from becoming victims of fraud in real life.

[0406] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0407] Step 1:

[0408] The server uses a generative AI model to generate a profile for the fraudster character. In this process, the server sends a prompt (e.g., "Male in his 20s, polite speaking, tech support fraud") as input to the generative AI model. The generative AI model creates a profile for the fraudster character based on this prompt, and generates detailed information about the character (such as age, gender, speech, and behavioral patterns) as output, which is then stored on the server.

[0409] Step 2:

[0410] The server uses the generative AI model to generate a dialogue scenario between the scammer character and the user. In this scenario generation process, the server inputs data about fraudulent methods (e.g., specific methods of phishing scams) and sends it to the generative AI model. The dialogue scenario is generated based on this input data, and detailed text data of the scenario is obtained as output. The server then sends this scenario data to the user's device.

[0411] Step 3:

[0412] The terminal presents the dialogue scenario sent from the server to the user. At this stage, the user views the scenario through the terminal and begins dialogue with the virtual impostor character. The user's responses are input by the terminal and transmitted to the server.

[0413] Step 4:

[0414] The device inputs video and audio data acquired from the user's camera and microphone into the emotion engine to recognize the user's real-time emotional state. The emotion engine analyzes this input data and outputs the user's emotional state (e.g., anxiety, alertness, relaxation). This emotional data is sent to the server.

[0415] Step 5:

[0416] The server collects user response data and emotion data and provides feedback to the generative AI model and emotion engine. During this feedback process, the server performs data analysis based on the collected data and outputs parameters to improve the generative AI model and emotion engine. The generative AI model and emotion engine use this feedback to improve future dialogue scenarios and emotion recognition accuracy.

[0417] Step 6:

[0418] The server uses an improved generative AI model and emotion engine to adjust the impostor character's responses in real time based on the user's emotional state. For example, if the user is feeling anxious, the generative AI model generates a reassuring response such as "Remote operation is very safe and we can solve the problem immediately" and sends it to the device. The device then presents this response to the user and continues the dialogue.

[0419] Through these steps, the system allows users to experience fraudulent tactics in real time and provides feedback according to their emotional state, effectively raising their vigilance.

[0420] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0421] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0422] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0423] [Second embodiment]

[0424] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0425] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0426] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0427] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0428] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0429] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0430] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0431] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0432] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0433] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0434] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0435] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0436] This invention relates to a system that uses a generative AI model to generate fraudster characters, and aims to improve users' understanding of fraudulent methods and their vigilance through dialogue between the users and the virtual fraudster characters. This system consists of a server, a terminal, and a user, and each component operates as follows:

[0437] Setting up the con man character

[0438] The server inputs conditions and parameters for generating a con artist character profile into the generative AI model. These conditions include the con artist's age, gender, speech characteristics, behavioral patterns, etc. The generative AI model uses these inputs to generate a specific con artist character profile, which is then stored on the server.

[0439] Interaction generation with virtual characters

[0440] A user accesses the system using a terminal and logs in to begin an impostor experience session. The terminal sends the user's authentication information to the server, which verifies it. If authentication is successful, the server uses a generative AI model to generate an interaction scenario with the impostor character. This scenario is displayed on the user's terminal, and the user begins interacting with it.

[0441] Recreating the fraud method

[0442] The server uses the generative AI model to generate scenarios for recreating specific fraud methods (e.g., phishing scams, technical support scams, etc.). The device presents the generated scenarios to the user in real time and receives the user's responses. This allows the user to interact with the fraudster character and understand the fraud methods and patterns.

[0443] Analysis and improvement of dialogue data

[0444] The server collects and stores dialogue data between users and virtual con artists. This data is fed back to the generative AI model, which analyzes the behavioral patterns of con artists and trends in fraudulent methods. Based on the analysis results, the generative AI model is improved. These improvements make future dialogue scenarios more realistic and effective, improving the accuracy of replicating fraudulent methods.

[0445] Specific examples

[0446] For example, suppose you want to create a male con artist character in his 20s. The server inputs the following conditions into the generative AI model: "male in his 20s, polite speaking, fraudulent technical support." Based on this, the generative AI model generates a specific con artist character. When a user accesses the system from a terminal and starts a con artist experience session, the following dialogue is generated:

[0447] Scammer: "Hello, this is tech support. Are you having trouble with your computer?"

[0448] User: "Yes, something seems to be wrong."

[0449] Scammer: "So, can you grant me remote access?"

[0450] Through these interactions, users can experience the techniques used by technical support scammers and become more vigilant against them. The server collects and analyzes the interaction data and improves the AI ​​model, enabling it to generate more accurate scammer characters and provide more accurate interaction scenarios.

[0451] The processing flow will be explained below.

[0452] Specific explanation of the program's processing steps

[0453] Step 1: Setting up your imposter character

[0454] 1. The server inputs the conditions and parameters for generating a profile of a con man character into the generative AI model.

[0455] Specifically, information such as the scammer's age, gender, language characteristics, and behavioral patterns is set.

[0456] 2. A generative AI model uses these inputs to generate a profile of a specific imposter character.

[0457] The profile includes the scammer's background, the fraudulent methods they use, and how they lead you to the scam.

[0458] 3. The server stores the generated profile in a database.

[0459] Step 2: Authenticating the user and starting an interactive session

[0460] 1. The user accesses the system from a terminal and logs in.

[0461] The user enters authentication information (e.g., username and password) through a dedicated interface.

[0462] 2. The terminal sends a login request to the server.

[0463] 3. The server checks the database and, if authentication is successful, allows the interactive session to begin.

[0464] If the authentication fails, an error message is returned to the terminal.

[0465] Step 3: Creating interactions with virtual characters

[0466] 1. The user operates the device and sends a request to start an "impostor experience session."

[0467] The user clicks the "Start Trial Session" button within the interface.

[0468] 2. The server uses the generative AI model to generate dialogue scenarios with the impostor character.

[0469] The generative AI model creates initial dialogue scenarios and response patterns based on the profile of the conman character.

[0470] 3. The terminal displays the dialogue scenario to the user in real time and receives the user's responses.

[0471] Responses from the con man character are generated and displayed sequentially in response to user input.

[0472] Step 4: Recreate the scam

[0473] 1. The server uses a generative AI model to generate scenarios to recreate specific fraud schemes.

[0474] The server tells the AI ​​model what type of fraud scheme it wants to replicate, such as a "phishing scam" or a "tech support scam."

[0475] 2. The generative AI model generates dialogue scenarios based on the specified fraudulent tactics.

[0476] The scenario includes specific steps on how the scammer will deceive the target.

[0477] 3. The terminal displays the generated fraud scenario to the user and conducts a dialogue in real time.

[0478] By following the scenario, users can simulate the process of fraudulent schemes.

[0479] Step 5: Collect and analyze interaction data

[0480] 1. The server collects and stores in a database the interaction data between the user and the virtual impostor character.

[0481] All messages and responses that occur during each interactive session are collected as logs.

[0482] 2. The server applies a data analysis algorithm to analyze the collected interaction data.

[0483] Identify trends in fraudster behavior and tactics.

[0484] Step 6: Improving the generative AI model

[0485] 1. Based on the analysis results, the server provides improvement feedback to update the algorithm of the generative AI model.

[0486] The algorithm of the generative AI model is updated based on new patterns and techniques, and these are reflected in the next dialogue generation.

[0487] 2. The server deploys the system to a new version, providing users with the latest fraud techniques and countermeasures.

[0488] The system is automatically updated to reflect the latest fraud techniques and countermeasures.

[0489] In this way, the system not only allows users to experience real interactions with scammers, thereby raising their vigilance, but also allows users to analyze the generated data to understand scammers' behavioral patterns and new techniques, further improving the generative AI model.

[0490] Example 1

[0491] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0492] There is a lack of systems to raise awareness and vigilance against real-world fraud methods. While the risk of being victimized increases when personal information is carelessly disclosed, there are limited ways to learn effective prevention measures without experiencing actual fraud methods. In addition, there is a need for technology that can quickly respond to fraud methods that are constantly evolving.

[0493] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0494] In this invention, the server includes means for generating a profile of a fraudster character using a generative AI model, means for generating a dialogue scenario between the fraudster character and a user, means for presenting the generated dialogue scenario and receiving the user's responses, means for collecting and analyzing the received dialogue data, means for improving the generative AI model based on the analyzed data, means for receiving and authenticating authentication information, and means for updating the dialogue scenario based on the user's responses. This allows users to safely and effectively experience and learn realistic fraud techniques. Furthermore, the continuous improvement process using the generative AI model provides practical learning that always keeps up with the latest fraud techniques.

[0495] A "generative AI model" is an artificial intelligence technology that learns from a variety of data and generates new characters and dialogue scenarios.

[0496] A "profile" is information that details the characteristics and attributes of a particular character.

[0497] A "scammer character" is a virtual character that imitates fraudulent methods and reproduces the fraudulent tactics through dialogue with users.

[0498] A "dialogue scenario" is a script that comprises a series of dialogues that take place between the con man character and the user.

[0499] "User" means any person or entity that uses the System.

[0500] "Authentication information" refers to information used to verify a user's identity, and includes, for example, a username and password.

[0501] "Received dialogue data" is information regarding the dialogue that took place between the user and the impostor character.

[0502] "Collection and analysis" is the process of gathering interaction data and examining its content and trends.

[0503] "Improving a generative AI model" is the process of improving the performance and accuracy of an AI model based on collected and analyzed data.

[0504] "Updating" means to generate a new dialogue scenario based on the user's response and continue the dialogue scenario.

[0505] "Authentication" is the process of verifying that a user has valid access rights.

[0506] The system of the present invention comprises a server, a terminal, and a user. A specific embodiment will be described below.

[0507] Setting up the con man character

[0508] The server first generates a profile for the con man character using a generative AI model. Parameters such as age, gender, speech characteristics, and behavioral patterns are used to generate the profile. Specifically, the server inputs the following prompt sentence into the generative AI model:

[0509] "Generate a male in his 20s, polite-spoken, fraudulent technical support scammer character."

[0510] The generative AI model generates a profile of the imposter character based on this prompt sentence and saves the profile on the server.

[0511] User Login and Authentication

[0512] A user accesses the system using a terminal and enters authentication information (e.g., username and password) on the login screen. The terminal then sends this authentication information to the server.

[0513] The server checks the received authentication information against the database, and if authentication is successful, allows the user access. If invalid authentication information is entered, an error message is returned to the terminal.

[0514] Start of the Impostor Experience Session

[0515] After the user is successfully authenticated, they send a request to the server to start an impostor experience session. The server then sends a request to the generative AI model to generate a dialogue scenario with the impostor character.

[0516] The generative AI model generates a dialogue scenario for the con man character and returns the scenario to the server, which then sends the generated scenario to the device.

[0517] Recreating the fraud method

[0518] The terminal presents the dialogue scenario received from the server to the user in real time, and the user inputs a response according to the dialogue scenario, which is then sent to the server by the terminal.

[0519] The server receives the user's response and requests the generative AI model to continue the dialogue based on that response. The generative AI model generates a new dialogue and returns it to the server. The server then sends this new dialogue to the device. This process is repeated, and the user experiences a dialogue with the virtual impostor character.

[0520] Analysis and improvement of dialogue data

[0521] The server collects and stores all interaction data between the user and the virtual impostor character, which is stored in a database for analysis.

[0522] The server uses the collected dialogue data to provide feedback to the generative AI model, improving the behavior of the conman character and the algorithm for generating dialogue scenarios, making future dialogue scenarios more realistic and effective.

[0523] Specific examples

[0524] For example, to create a male con man character in his 20s, the server inputs conditions such as "male in his 20s, polite speech, fraudulent technical support" into the generative AI model. The generative AI model generates a specific con man character based on these conditions. When a user accesses the system from a terminal and starts a con man experience session, the following dialogue is generated:

[0525] Scammer: "Hello, this is tech support. Are you having trouble with your computer?"

[0526] User: "Yes, something seems to be wrong."

[0527] Scammer: "So, can you grant me remote access?"

[0528] Through these interactions, users can experience tech support scams and become more vigilant against them. The server collects and analyzes the interaction data and improves the generative AI model, enabling it to generate more accurate scammer characters and provide more accurate interaction scenarios.

[0529] Summary of prompt sentence examples

[0530] "Generate a male in his 20s, polite-spoken, fraudulent technical support scammer character."

[0531] "Generate a phishing scenario"

[0532] "Continue the dialogue of the imposter character based on the user's response."

[0533] This will enable users to learn specific countermeasures against various fraudulent methods and create a system that can prevent fraud damage before it occurs.

[0534] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0535] Step 1:

[0536] The server inputs a prompt sentence for generating a profile to the generative AI model.

[0537] Input: Character attribute information (e.g., age, gender, speech characteristics, behavior patterns)

[0538] Data processing: The generative AI model generates a character profile based on the input attribute information.

[0539] Output: Generated character profile

[0540] Specific operation: For example, a prompt such as "Male in his 20s, polite speaking, fraudulent technical support" is sent to the generation AI model, and a profile of the fraudster character is generated and stored on the server.

[0541] Step 2:

[0542] A user accesses the system using a terminal and enters authentication information (e.g., username and password) on a login screen.

[0543] Input: User authentication information (username, password)

[0544] Data processing: The terminal sends the entered authentication information to the server.

[0545] Output: Sending authentication information to the server

[0546] Specific operation: The user enters the username and password on the login screen and clicks the "Login" button.

[0547] Step 3:

[0548] The server checks the received authentication information against a database and allows the user access if authentication is successful.

[0549] Input: User authentication information sent from the device

[0550] Data processing: Check against the database to determine whether authentication was successful.

[0551] Output: Permissions or error messages

[0552] Specific operation: The server checks the authentication information stored in the database, and if it matches, notifies the terminal that authentication was successful.

[0553] Step 4:

[0554] After the user is successfully authenticated, the terminal sends a request to the server to initiate an impostor experience session.

[0555] Input: Session initiation request

[0556] Data processing: The server receives the request and begins processing.

[0557] Output: Session start confirmation message

[0558] Specific action: The user clicks the "Start Impostor Experience Session" button and submits the request.

[0559] Step 5:

[0560] The server sends a request to the generative AI model to generate an interaction scenario with the impostor character.

[0561] Input: Impostor Experience Session Start Request

[0562] Data processing: The generative AI model generates dialogue scenarios with the con man character.

[0563] Output: Generated dialogue scenario

[0564] Specific operation: The generative AI model generates a "tech support scam" scenario and returns it to the server.

[0565] Step 6:

[0566] The terminal presents the dialogue scenario received from the server to the user in real time.

[0567] Input: Interaction scenario sent from the server

[0568] Data processing: The scenario received by the terminal is displayed on the screen.

[0569] Output: Display of the interaction scenario to the user

[0570] What it does: The device displays the scammer character's line, "Hello, this is tech support. Are you having trouble with your computer?"

[0571] Step 7:

[0572] The user inputs a response according to the dialogue scenario, and the response is transmitted by the terminal to the server.

[0573] Input: User response (e.g. "Yes, something seems to be wrong.")

[0574] Data processing: The terminal sends the user's response to the server.

[0575] Output: Sending the user response to the server

[0576] Specific behavior: The user types a response and clicks the send button.

[0577] Step 8:

[0578] The server receives the user's response and, based on that, requests the generative AI model to continue the dialogue.

[0579] Input: User response

[0580] Data processing: Ask the generative AI model to continue the dialogue and generate a new scenario.

[0581] Output: Generated new dialogue scenario

[0582] Specific behavior: The generative AI model generates a new dialogue scenario and generates the reply, "So, may I grant you remote access?"

[0583] Step 9:

[0584] The server transmits the generated new dialogue scenario to the terminal.

[0585] Input: Generated new dialogue scenario

[0586] Data processing: Send the dialogue scenario to the terminal.

[0587] Output: Sending the dialogue scenario to the terminal

[0588] Specific operation: The server sends the newly generated dialogue scenario to the terminal, which then displays it to the user.

[0589] Step 10:

[0590] The server collects, stores, and analyzes all interaction data between the user and the virtual impostor character.

[0591] Input: Interaction data between the user and the impostor character

[0592] Data processing: Collection, storage, and analysis of interaction data

[0593] Output: Analysis results

[0594] What it does: The server stores the collected interaction data in a database and uses data analysis tools to extract patterns.

[0595] Step 11:

[0596] The server provides feedback to the generative AI model based on the analysis results, improving the AI ​​model.

[0597] Input: Analysis results

[0598] Data processing: Updating generative AI models

[0599] Output: An improved generative AI model

[0600] How it works: The generative AI model is refined based on newly collected data and analysis results, improving the quality of future dialogue scenarios.

[0601] (Application example 1)

[0602] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0603] Fraud methods continue to become more sophisticated, increasing the risk of many people becoming victims of fraud. Fraud methods via the Internet and telephone are particularly complex, making it difficult for many people to experience the risks in a realistic manner and raise their vigilance. Therefore, there is a need for a system that allows users to experience actual fraud methods and raise their vigilance.

[0604] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0605] In this invention, the server includes means for generating a profile of a fraudster character using a generative AI model, means for generating a dialogue scenario between the fraudster character and a user, and means for providing the generated dialogue scenario in a smartphone application, thereby enabling users to experience fraudulent tactics in a realistic manner through their smartphones and increase their vigilance.

[0606] A "generative AI model" is an artificial intelligence model that extracts specific patterns and information from data and generates natural language.

[0607] A "scammer character" is a profile of a virtual scammer created by a generative AI model, including age, gender, language, and behavioral patterns.

[0608] A "dialogue scenario" is a script that shows the flow of a pre-set conversation that takes place between a con man character and a user.

[0609] A "smartphone application" is a part of a program that runs on a smartphone and provides an interface for a user to interact with an impostor character.

[0610] "User response" refers to a reply or action that a user inputs or selects in a dialogue scenario.

[0611] "Dialogue data" is a record of the dialogue that took place between the user and the impostor character.

[0612] "Improving the generative AI model" refers to the process of analyzing collected interaction data and using the results to further improve the generative AI model.

[0613] A "fraudulent technique" is a specific method or means used by a fraudster to deceive a victim.

[0614] "Interactive experience to increase vigilance" refers to an educational dialogue that allows users to virtually experience fraudulent methods in order to increase their vigilance against real-life scams.

[0615] This invention relates to a system that uses a generative AI model to generate fraudster characters, allowing users to virtually experience fraud and raise their vigilance. The system consists of a server, a terminal (such as a smartphone), and a user.

[0616] Server Processing

[0617] The server uses a generative AI model to generate a profile of the fraudster character. To generate the profile, the user inputs conditions such as the fraudster's age, gender, speech characteristics, and behavioral patterns. This generates a specific fraudster character and stores it on the server.

[0618] The server then uses the generative AI model to generate a dialogue scenario for the user. This dialogue scenario is generated based on a specific fraudulent scheme (e.g., technical support fraud or phishing scam). The generated scenario is saved on the server and then sent to the user's device (smartphone).

[0619] Terminal (smartphone) processing

[0620] A user accesses the system using a smartphone and logs in. After logging in, the device sends the user's authentication information to the server, and authentication is performed. If authentication is successful, the server sends a dialogue scenario with the impostor character to the device and presents it to the user.

[0621] When the user initiates a dialogue, the device receives the user's response and sends it to the server. As the dialogue between the user and the impostor character progresses, the device continues to update the scenario in real time.

[0622] User Experience

[0623] The user experiences a virtual fraud through a smartphone application. For example, if the user selects the "tech support fraud" scenario, the following dialogue is generated:

[0624] Scammer Character: "Hello, tech support. Having trouble with your computer?"

[0625] User: "Yes, something seems to be wrong."

[0626] Impostor Character: "So, can you grant me remote access?"

[0627] Through such interactions, users can develop a sense of vigilance against real scams.

[0628] Data collection and analysis

[0629] Once the conversation ends, the server collects and stores the conversation data with the user. This data is fed back to the generative AI model, which analyzes the behavioral patterns of fraudsters and trends in fraudulent methods. Based on the analysis results, the generative AI model is improved. These improvements make future conversation scenarios more realistic and effective, improving the accuracy of replicating fraudulent methods.

[0630] Prompt Sentence Examples

[0631] Here's an example of a specific prompt for a generative AI model to generate a profile for a con man character:

[0632] "Male, 25 years old, polite speaking, fraudulent tech support scam. Generate a scenario to convince a user to grant remote access."

[0633] The present invention provides an effective means for users to experience virtual fraud through a smartphone application, thereby increasing their vigilance against real-life fraud.

[0634] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0635] Step 1:

[0636] The server inputs a prompt to the generative AI model to generate a profile of a con man character. This prompt includes the con man's age, gender, language, and behavioral patterns. Based on these conditions, the generative AI model generates a profile of a specific con man character and saves the profile on the server. The input data is the prompt, and the output data is the con man character's profile.

[0637] Step 2:

[0638] The server uses the generative AI model to generate a dialogue scenario between the conman character and the user. Specifically, the server inputs data on the conman character's profile and fraudulent methods into the generative AI model, and outputs a dialogue scenario. The generated dialogue scenario is saved on the server. The input data is the conman character's profile and fraudulent methods, and the output data is the dialogue scenario.

[0639] Step 3:

[0640] A user accesses the system using a smartphone terminal and logs in. The terminal sends the user's authentication information (username and password) to the server, which then authenticates it. If authentication is successful, the server sends the generated dialogue scenario to the terminal. The input data is the user's authentication information, and the output data is the authentication result and the dialogue scenario.

[0641] Step 4:

[0642] When the user selects to start the scenario, the smartphone device begins a dialogue with the imposter character. The user inputs a response to the imposter character's prompt, which is then sent from the device to the server. The server receives the user's response and updates the dialogue scenario as necessary. The input data is the user's response, and the output data is the updated dialogue scenario.

[0643] Step 5:

[0644] Once the interaction is complete, the server collects and stores the interaction data with the user. This interaction data includes the user's responses and the statements made by the imposter character. The server analyzes this data and feeds it back into the generative AI model. The input data is the interaction data, and the output data is the analysis results.

[0645] Step 6:

[0646] The server improves the generative AI model based on the analysis results. This makes the generation of future dialogue scenarios more realistic and effective, improving the accuracy of reproducing fraudulent methods. The input data is the analysis results, and the output data is the improved generative AI model.

[0647] Through the above processing steps, users can virtually experience fraud through a smartphone application, increasing their vigilance against real-life fraud.

[0648] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0649] This invention relates to a system that generates fraudster characters by combining a generative AI model and an emotion engine, and aims to improve users' understanding of fraudulent methods and their vigilance through dialogue between the user and the virtual fraudster character. This system consists of a server, a terminal, a user, and an emotion engine, and each component operates as follows:

[0650] Setting up the con man character

[0651] The server inputs conditions and parameters for generating a con artist character profile into the generative AI model. These conditions include the con artist's age, gender, speech characteristics, behavioral patterns, etc. The generative AI model uses these inputs to generate a specific con artist character profile, which is then stored on the server.

[0652] Authenticating the user and starting an interactive session

[0653] A user accesses the system using a terminal and logs in to begin an impostor experience session. The terminal sends the user's authentication information to the server, which verifies it. If authentication is successful, the server uses a generative AI model to generate an interaction scenario with the impostor character. This scenario is displayed on the user's terminal, and the user begins interacting with it.

[0654] Incorporating an emotion engine

[0655] The emotion engine analyzes video and audio data acquired from the user's camera and microphone to recognize the user's real-time emotional state. This emotional data is sent to the server and reflected in the responses of the con man character and the progression of the scenario.

[0656] Fraud reenactment and emotional feedback

[0657] The server uses the generative AI model to generate scenarios for recreating specific fraud schemes (e.g., phishing scams, technical support scams, etc.). The device presents the generated scenarios to the user in real time, and the emotion engine analyzes the user's emotional state. This allows the fraudster character to respond appropriately based on the user's emotional data. For example, if the user is feeling anxious, the fraudster character will use words that will further reassure them.

[0658] Collection and analysis of dialogue data

[0659] The server collects dialogue and emotional data between the user and the virtual con man character and stores it in a database. This data is fed back to the generative AI model and emotion engine, which analyzes trends in the con man's behavioral patterns and emotional responses. Based on the analysis results, both models are improved. These improvements make future dialogue scenarios more realistic and effective, improving the accuracy of reproducing fraudulent methods.

[0660] Specific examples

[0661] For example, let's create a male scammer character in his 20s on the theme of technical support fraud. The server inputs the following conditions into the generative AI model: "male in his 20s, polite speech, fraudulent technical support." Based on this, the generative AI model generates a specific scammer character. When a user accesses the system from a terminal and starts a scammer experience session, the following dialogue is generated:

[0662] Scammer: "Hello, this is tech support. Are you having trouble with your computer?"

[0663] User: "Yes, something seems to be wrong."

[0664] Scammer: "So, can you grant me remote access?"

[0665] During this dialogue, the emotion engine analyzes the user's facial expressions and voice to recognize their emotions in real time. If the emotion engine determines that the user is feeling anxious, the con man character will respond with a reassuring message, saying, "Remote operation is very safe and we can resolve the issue immediately." The server collects dialogue data and emotion data and feeds it back to the generative AI model and emotion engine to improve the accuracy of the system.

[0666] In this way, the system not only allows users to experience real interactions with scammers, thereby raising their vigilance, but also analyzes the generated data to understand scammers' behavioral patterns and new techniques, further improving the generative AI model and emotion engine.

[0667] The processing flow will be explained below.

[0668] Specific explanation of the program's processing steps

[0669] Step 1: Setting up your imposter character

[0670] 1. The server inputs the conditions and parameters for generating a profile of a con man character into the generative AI model.

[0671] The criteria include age, gender, language characteristics, and behavioral patterns.

[0672] 2. The generative AI model uses these conditions to generate a profile of a specific imposter character.

[0673] The profile includes the scammer's background, the fraudulent methods they use, and their luring techniques.

[0674] 3. The server stores the generated profile in a database.

[0675] Step 2: Authenticating the user and starting an interactive session

[0676] 1. A user accesses the system using a terminal and logs in.

[0677] The user enters authentication information (user name, password) in a dedicated interface.

[0678] 2. The terminal sends a login request to the server.

[0679] 3. The server checks the database and, if authentication is successful, allows the interactive session to begin.

[0680] If the authentication fails, an error message is returned to the terminal.

[0681] Step 3: Creating interactions with virtual characters

[0682] 1. The user operates the device and sends a request to the server to start a "scammer experience session."

[0683] The user clicks the "Start Trial Session" button within the interface.

[0684] 2. The server uses the generative AI model to generate dialogue scenarios with the impostor character.

[0685] The generative AI model creates initial dialogue scenarios and response patterns based on the profile of the conman character.

[0686] 3. The terminal displays the dialogue scenario to the user in real time and receives the user's responses.

[0687] Responses from the con man character are generated and displayed sequentially in response to user input.

[0688] Step 4: Replaying the scam and emotional feedback

[0689] 1. The server uses a generative AI model to generate scenarios to recreate specific fraud schemes.

[0690] The server instructs the AI ​​model on the type of fraudulent scheme it wants to replicate, such as "phishing scams" or "tech support scams."

[0691] 2. The generative AI model generates dialogue scenarios based on the specified fraudulent tactics.

[0692] The scenarios include specific steps on how the scammer will deceive the target.

[0693] 3. The emotion engine analyzes video and audio data obtained from the user's camera and microphone to recognize the user's emotional state.

[0694] Emotions are classified as, for example, "relief," "anxiety," and "doubt."

[0695] 4. The device displays the generated fraud scenario to the user, and the emotion engine analyzes the user's emotional state and transmits it to the server in real time.

[0696] The server then adjusts the imposter character's responses based on this: if the user is feeling anxious, the imposter character will respond in a reassuring way.

[0697] Step 5: Collect and analyze interaction data

[0698] 1. The server collects and stores in a database the conversation data and emotion data exchanged between the user and the impostor character.

[0699] All messages and emotional responses from each interaction session are logged.

[0700] 2. The server applies data analysis algorithms to analyze the collected interaction data and emotion data.

[0701] Identify trends in fraudster behavior patterns and user emotional responses.

[0702] Step 6: Improving the generative AI model and emotion engine

[0703] 1. Based on the analysis results, the server provides feedback to update the generative AI model algorithm and the emotion engine analysis.

[0704] The algorithm of the generative AI model is updated based on new patterns and techniques, and is reflected in the next dialogue generation. The emotion engine is also updated to improve the accuracy of the user's emotional response.

[0705] 2. The server deploys the system to a new version and provides users with an interactive experience that helps them understand and be aware of the latest fraud techniques.

[0706] The system is automatically updated to reflect the latest fraud techniques and countermeasures.

[0707] In this way, the system not only allows users to experience real-life interactions with scammers, thereby raising their vigilance, but also analyzes the generated data to understand scammers' behavioral patterns and new techniques, further improving the generative AI model and emotion engine.

[0708] Example 2

[0709] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0710] With the advancement of modern advanced communication and information technologies, fraud methods are becoming more sophisticated. In particular, fraud methods that skillfully exploit users' psychological state can cause serious damage. However, many users are not sufficiently vigilant against fraudulent methods and are easily deceived. Therefore, effective measures are needed to help users better understand fraudulent methods and increase their vigilance.

[0711] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0712] In this invention, the server includes means for generating a profile of a fraudster character using a generative AI model, means for generating a dialogue scenario between the fraudster character and the user, means for presenting the generated dialogue scenario to the user and receiving the user's response, means for acquiring real-time emotional data of the user using an emotion engine that recognizes the user's emotional state, means for collecting and analyzing the received dialogue data and emotional data, and means for improving the generative AI model based on the analyzed data. This allows the user to learn fraudulent tactics through dialogue with the fraudster character and to increase their vigilance against real-world fraud by receiving feedback on their emotional state in real time.

[0713] A "generative AI model" is a model that uses artificial intelligence technology to generate characters and dialogue scenarios tailored to specific purposes.

[0714] A "scammer character" is an artificially generated character that simulates fraudulent methods, and allows the user to learn fraudulent techniques and methods through dialogue with the user.

[0715] A "profile" is a set of information that includes characteristics such as the conman character's age, gender, speech patterns, and behavioral patterns.

[0716] A "dialogue scenario" specifically defines the flow and content of a series of dialogues that take place between a con man character and a user.

[0717] The "emotion engine" is a system that analyzes the user's facial expressions and voice through devices such as cameras and microphones, and recognizes their emotional state at that time in real time.

[0718] "Real-time" refers to reactions and analysis that occur immediately on the spot while the user is using the system.

[0719] "Emotion data" is information about the emotional state obtained from the user's facial expressions and voice analyzed by the emotion engine.

[0720] "Dialogue data" is information about the content of the conversation between the user and the impostor character, including the flow of the dialogue and specific wording.

[0721] "Analysis" refers to the process of using collected interaction and emotion data to evaluate patterns and trends in order to improve the performance of generative AI models and emotion engines.

[0722] This invention relates to a system that combines a generative AI model and an emotion engine to generate a fraudster character, and aims to improve understanding of fraudulent methods and vigilance through dialogue between the user and the virtual fraudster character. This system consists of a server, a terminal, a user, and an emotion engine, and each component operates as follows:

[0723] Setting up the con man character

[0724] The server inputs the conditions and parameters for generating a scammer character profile into the generative AI model. These conditions include the scammer's age, gender, speech characteristics, and behavioral patterns. For example, a prompt might read, "Please generate a character who is a man in his 20s who speaks politely and commits technical support scams." Based on these inputs, the generative AI model generates a profile for a specific scammer character, and the profile is stored on the server.

[0725] Authenticating the user and starting an interactive session

[0726] A user accesses the system using a terminal and logs in to begin an imposter experience session. The terminal sends the user's authentication information to the server, which verifies it. If authentication is successful, the server uses the generative AI model to generate an interaction scenario with the imposter character. For example, this can be achieved by sending the generative AI model a prompt such as "Please generate a tech support fraud scenario." This scenario is then displayed on the user's terminal, and the user begins interacting with it.

[0727] Incorporating an emotion engine

[0728] The emotion engine analyzes video and audio data acquired from the user's camera and microphone to recognize the user's real-time emotional state. This emotional data is sent to the server and reflected in the responses of the con man character and the progression of the scenario.

[0729] Fraud reenactment and emotional feedback

[0730] The server uses the generative AI model to generate scenarios for recreating specific fraud schemes (e.g., phishing scams, technical support scams, etc.). The device presents the generated scenarios to the user in real time, and the emotion engine analyzes the user's emotional state. This allows the fraudster character to respond appropriately based on the user's emotional data. For example, if the user is feeling anxious, the fraudster character will use words that will further reassure them.

[0731] Collection and analysis of dialogue data

[0732] The server collects dialogue and emotional data between the user and the virtual con man character and stores it in a database. This data is fed back to the generative AI model and emotion engine, which analyzes trends in the con man's behavioral patterns and emotional responses. Based on the analysis results, both models are improved. These improvements make future dialogue scenarios more realistic and effective, improving the accuracy of reproducing fraudulent methods.

[0733] Specific examples

[0734] For example, if you create a male scammer character in his 20s on the theme of technical support fraud, the server inputs the following conditions into the generative AI model: "male in his 20s, politely speaking, technical support fraud." Based on this, the generative AI model generates a specific scammer character.

[0735] When a user accesses the system from a terminal and starts an imposter experience session, the following dialogue is generated:

[0736] Scammer: "Hello, this is tech support. Are you having trouble with your computer?"

[0737] User: "Yes, something seems to be wrong."

[0738] Scammer: "So, can you grant me remote access?"

[0739] During this dialogue, the emotion engine analyzes the user's facial expressions and voice to recognize their emotions in real time. For example, if the emotion engine determines that the user is feeling anxious, the con artist character will respond with a reassuring "Remote operation is very safe, and we can resolve the issue immediately." The server collects dialogue and emotion data and feeds it back to the generative AI model and emotion engine to improve the system's accuracy. In this way, users not only experience realistic dialogue with con artists and become more vigilant, but also analyze the generated data to understand the con artists' behavioral patterns and new techniques, allowing them to make future dialogue scenarios more realistic and effective.

[0740] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0741] Step 1:

[0742] Setting up the con man character

[0743] The server inputs conditions and parameters for generating a profile of a con man character into the generative AI model. The input conditions include the con man's age, gender, speech characteristics, and behavioral patterns. The generative AI model then generates a profile of the con man character based on those conditions. The generated profile is stored in the server's database.

[0744] Specific operation: The server sends the prompt "Generate a character who is a man in his 20s and who will commit technical support scams in a polite tone" to the generation AI model and saves the profile of the resulting scammer character.

[0745] Input: Criteria for the conman character (age, gender, speech characteristics, behavioral patterns)

[0746] Output: Impostor character profile

[0747] Step 2:

[0748] Authenticating Users

[0749] A user accesses the system using a terminal and enters authentication information on the login screen. The terminal sends the authentication information to the server. The server compares the received authentication information with a database and determines whether the authentication is successful or not. If the authentication is successful, the server sends a session start notification to the user.

[0750] How it works: The user enters their username and password on the login screen, and the device sends this to the server, which then checks the information against the database and returns the results.

[0751] Input: User authentication information (username, password)

[0752] Output: Authentication result (success or failure)

[0753] Step 3:

[0754] Start of the Impostor Experience Session

[0755] The server generates a dialogue scenario with the impostor character using the generative AI model. The generated dialogue scenario is sent to the user's device and displayed. The user then begins a dialogue with the impostor character based on the scenario.

[0756] Specific operation: The server sends a prompt to the generative AI model saying, "Please generate a technical support fraud scenario," and presents the resulting scenario to the user's device.

[0757] Input: Prompt sentence (instruction to generate a dialogue scenario with the impostor character)

[0758] Output: Generated dialogue scenario

[0759] Step 4:

[0760] Incorporating an emotion engine

[0761] The device captures video and audio data from the user's camera and microphone. The emotion engine analyzes this data and recognizes the user's emotional state in real time. This emotional data is then sent to the server.

[0762] Specific operation: The device sends video and audio data to the emotion engine, which analyzes it to generate real-time emotion data and send it to the server.

[0763] Input: Video data, audio data

[0764] Output: User emotion data

[0765] Step 5:

[0766] Fraud reenactment and emotional feedback

[0767] The server uses the generative AI model to generate scenarios for recreating specific fraud schemes, which are then presented to the user, and the responses of the fraudster character are adjusted based on data from the emotion engine.

[0768] Specific operation: The server sends a prompt to the generative AI model to "generate a scenario that reproduces a fraudulent scheme," and presents the generated scenario to the user in real time. Based on the emotional data, the fraudster character's responses are appropriately adjusted.

[0769] Input: prompt sentence (instructions for reproducing the fraud scheme), user emotion data

[0770] Output: Adjusted impostor character response

[0771] Step 6:

[0772] Collection and analysis of dialogue data

[0773] The server collects and stores the conversation and emotion data between the user and the impostor character in a database. The collected data is fed back to the generative AI model and emotion engine, which then improves both models.

[0774] How it works: The server collects dialogue and emotion data in real time, stores it in a database, analyzes it, and uses it as feedback to improve the performance of the generative AI model and emotion engine.

[0775] Input: Dialogue data, emotion data

[0776] Output: Improved generative AI model, emotion engine

[0777] (Application example 2)

[0778] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0779] In modern society, fraudsters' tactics are becoming increasingly sophisticated, increasing the risk of ordinary users being scammed. Therefore, there is a need for effective training methods to help users understand actual fraudulent methods and increase their vigilance. However, current technology makes it difficult for users to interactively experience realistic fraud scenarios and receive feedback based on their emotional state while increasing their vigilance. To solve this problem, the development of an innovative system that combines a generative AI model and an emotion engine is desired.

[0780] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0781] In this invention, the server includes: means for generating a profile of a fraudster character using a generative AI model; means for generating a dialogue scenario between the fraudster character and a user; means for presenting the generated dialogue scenario to the user and receiving the user's response; means for collecting and analyzing the received dialogue data and the user's emotional data; means for improving the generative AI model and the emotion engine based on the analyzed data; and means including an emotion engine for adjusting the dialogue scenario based on the user's emotional data. This allows the user to experience fraudulent techniques in real time and receive appropriate feedback according to their emotional state, effectively increasing their vigilance against fraud.

[0782] A "generative AI model" is a form of generative artificial intelligence (AI) that is an algorithm that learns from large amounts of data and generates new profiles or text based on specific conditions and parameters.

[0783] A "scammer character" is a virtual scammer profile constructed using a generative AI model, and is a person with specific attributes such as age, gender, language, and behavioral patterns.

[0784] A "dialogue scenario" refers to the sequence or story of a dialogue that takes place between a con man character constructed by a generative AI model and the user.

[0785] An "emotion engine" is a general term for algorithms and software that analyze a user's facial expressions, voice, actions, etc. in real time and recognize their emotional state.

[0786] "Emotional Data" refers to information about a user's emotional state obtained and analyzed by the emotion engine.

[0787] "Dialogue data" refers to the record of the dialogue between the user and the virtual con man character and the text data thereof.

[0788] A "profile" refers to detailed information about a person generated based on specific conditions or parameters, including, for example, age, gender, speech, and behavioral patterns.

[0789] "Real-time" refers to processing and responses occurring almost immediately, without delay, and indicates a state in which dialogue with the user and emotional analysis are carried out immediately.

[0790] "Feedback" refers to the return of information that allows a generative AI model or emotion engine to change its response or improve the system based on the user's reaction or emotional state.

[0791] The present invention is a system that combines a generative AI model and an emotion engine to help users understand fraudulent tactics and raise their vigilance through dialogue with virtual fraudster characters. The system has the following specific configuration and operation procedures:

[0792] Server Roles

[0793] The server uses a generative AI model to generate a profile for the scammer character by inputting prompt statements containing criteria such as "male in his 20s, polite speech, technical support scam" into the model.

[0794] example:

[0795] text

[0796] Create a scammer character with the following attributes: 20-year-old male with polite speech, specializing in technical support scams.

[0797] Furthermore, the server uses the generative AI model to generate dialogue scenarios between the scammer character and the user, based on scamming techniques (e.g., phishing scams and tech support scams).

[0798] Device Role

[0799] The terminal used by the user is a device such as a smartphone or a head-mounted display. The terminal presents the dialogue scenario sent from the server to the user and receives the user's response.

[0800] The device also inputs video and audio data acquired from the user's camera and microphone into an emotion engine to recognize the user's real-time emotional state. The emotion engine uses software such as Affectiva. The emotion data is sent to the server and reflected in the responses of the imposter character and the progression of the scenario. For example, if the user shows signs of anxiety, the imposter character will respond reassuringly, saying, "Remote operation is very safe and we can solve the problem immediately."

[0801] User Roles

[0802] Users access the system via a terminal, log in, and begin a fraudster experience session. During the session, users interact with a virtual fraudster character and respond according to the progression of the dialogue scenario. Through this interaction, users can experience realistic fraudulent tactics and become more vigilant.

[0803] The server collects and analyzes user response and emotion data. The analyzed data is fed back to the generative AI model and emotion engine. This further improves the generative AI model and emotion engine, making future dialogue scenarios more realistic and effective.

[0804] These configurations and operational procedures allow users to experience fraudulent techniques in real time while receiving appropriate feedback based on their emotional state, which can increase their vigilance against fraud and help prevent them from becoming victims of fraud in real life.

[0805] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0806] Step 1:

[0807] The server uses a generative AI model to generate a profile for the fraudster character. In this process, the server sends a prompt (e.g., "Male in his 20s, polite speaking, tech support fraud") as input to the generative AI model. The generative AI model creates a profile for the fraudster character based on this prompt, and generates detailed information about the character (such as age, gender, speech, and behavioral patterns) as output, which is then stored on the server.

[0808] Step 2:

[0809] The server uses the generative AI model to generate a dialogue scenario between the scammer character and the user. In this scenario generation process, the server inputs data about fraudulent methods (e.g., specific methods of phishing scams) and sends it to the generative AI model. The dialogue scenario is generated based on this input data, and detailed text data of the scenario is obtained as output. The server then sends this scenario data to the user's device.

[0810] Step 3:

[0811] The terminal presents the dialogue scenario sent from the server to the user. At this stage, the user views the scenario through the terminal and begins dialogue with the virtual impostor character. The user's responses are input by the terminal and transmitted to the server.

[0812] Step 4:

[0813] The device inputs video and audio data acquired from the user's camera and microphone into the emotion engine to recognize the user's real-time emotional state. The emotion engine analyzes this input data and outputs the user's emotional state (e.g., anxiety, alertness, relaxation). This emotional data is sent to the server.

[0814] Step 5:

[0815] The server collects user response data and emotion data and provides feedback to the generative AI model and emotion engine. During this feedback process, the server performs data analysis based on the collected data and outputs parameters to improve the generative AI model and emotion engine. The generative AI model and emotion engine use this feedback to improve future dialogue scenarios and emotion recognition accuracy.

[0816] Step 6:

[0817] The server uses an improved generative AI model and emotion engine to adjust the impostor character's responses in real time based on the user's emotional state. For example, if the user is feeling anxious, the generative AI model generates a reassuring response such as "Remote operation is very safe and we can solve the problem immediately" and sends it to the device. The device then presents this response to the user and continues the dialogue.

[0818] Through these steps, the system allows users to experience fraudulent tactics in real time and provides feedback according to their emotional state, effectively raising their vigilance.

[0819] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0820] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0821] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0822] [Third embodiment]

[0823] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0824] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0825] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0826] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0827] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0828] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0829] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0830] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0831] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0832] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0833] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0834] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[0835] This invention relates to a system that uses a generative AI model to generate fraudster characters, and aims to improve users' understanding of fraudulent methods and their vigilance through dialogue between the users and the virtual fraudster characters. This system consists of a server, a terminal, and a user, and each component operates as follows:

[0836] Setting up the con man character

[0837] The server inputs conditions and parameters for generating a con artist character profile into the generative AI model. These conditions include the con artist's age, gender, speech characteristics, behavioral patterns, etc. The generative AI model uses these inputs to generate a specific con artist character profile, which is then stored on the server.

[0838] Interaction generation with virtual characters

[0839] A user accesses the system using a terminal and logs in to begin an impostor experience session. The terminal sends the user's authentication information to the server, which verifies it. If authentication is successful, the server uses a generative AI model to generate an interaction scenario with the impostor character. This scenario is displayed on the user's terminal, and the user begins interacting with it.

[0840] Recreating the fraud method

[0841] The server uses the generative AI model to generate scenarios for recreating specific fraud methods (e.g., phishing scams, technical support scams, etc.). The device presents the generated scenarios to the user in real time and receives the user's responses. This allows the user to interact with the fraudster character and understand the fraud methods and patterns.

[0842] Analysis and improvement of dialogue data

[0843] The server collects and stores dialogue data between users and virtual con artists. This data is fed back to the generative AI model, which analyzes the behavioral patterns of con artists and trends in fraudulent methods. Based on the analysis results, the generative AI model is improved. These improvements make future dialogue scenarios more realistic and effective, improving the accuracy of replicating fraudulent methods.

[0844] Specific examples

[0845] For example, suppose you want to create a male con artist character in his 20s. The server inputs the following conditions into the generative AI model: "male in his 20s, polite speaking, fraudulent technical support." Based on this, the generative AI model generates a specific con artist character. When a user accesses the system from a terminal and starts a con artist experience session, the following dialogue is generated:

[0846] Scammer: "Hello, this is tech support. Are you having trouble with your computer?"

[0847] User: "Yes, something seems to be wrong."

[0848] Scammer: "So, can you grant me remote access?"

[0849] Through these interactions, users can experience the techniques used by technical support scammers and become more vigilant against them. The server collects and analyzes the interaction data and improves the AI ​​model, enabling it to generate more accurate scammer characters and provide more accurate interaction scenarios.

[0850] The processing flow will be explained below.

[0851] Specific explanation of the program's processing steps

[0852] Step 1: Setting up your imposter character

[0853] 1. The server inputs the conditions and parameters for generating a profile of a con man character into the generative AI model.

[0854] Specifically, information such as the scammer's age, gender, language characteristics, and behavioral patterns is set.

[0855] 2. A generative AI model uses these inputs to generate a profile of a specific imposter character.

[0856] The profile includes the scammer's background, the fraudulent methods they use, and how they lead you to the scam.

[0857] 3. The server stores the generated profile in a database.

[0858] Step 2: Authenticating the user and starting an interactive session

[0859] 1. The user accesses the system from a terminal and logs in.

[0860] The user enters authentication information (e.g., username and password) through a dedicated interface.

[0861] 2. The terminal sends a login request to the server.

[0862] 3. The server checks the database and, if authentication is successful, allows the interactive session to begin.

[0863] If the authentication fails, an error message is returned to the terminal.

[0864] Step 3: Creating interactions with virtual characters

[0865] 1. The user operates the device and sends a request to start an "impostor experience session."

[0866] The user clicks the "Start Trial Session" button within the interface.

[0867] 2. The server uses the generative AI model to generate dialogue scenarios with the impostor character.

[0868] The generative AI model creates initial dialogue scenarios and response patterns based on the profile of the conman character.

[0869] 3. The terminal displays the dialogue scenario to the user in real time and receives the user's responses.

[0870] Responses from the con man character are generated and displayed sequentially in response to user input.

[0871] Step 4: Recreate the scam

[0872] 1. The server uses a generative AI model to generate scenarios to recreate specific fraud schemes.

[0873] The server tells the AI ​​model what type of fraud scheme it wants to replicate, such as a "phishing scam" or a "tech support scam."

[0874] 2. The generative AI model generates dialogue scenarios based on the specified fraudulent tactics.

[0875] The scenario includes specific steps on how the scammer will deceive the target.

[0876] 3. The terminal displays the generated fraud scenario to the user and conducts a dialogue in real time.

[0877] By following the scenario, users can simulate the process of fraudulent schemes.

[0878] Step 5: Collect and analyze interaction data

[0879] 1. The server collects and stores in a database the interaction data between the user and the virtual impostor character.

[0880] All messages and responses that occur during each interactive session are collected as logs.

[0881] 2. The server applies a data analysis algorithm to analyze the collected interaction data.

[0882] Identify trends in fraudster behavior and tactics.

[0883] Step 6: Improving the generative AI model

[0884] 1. Based on the analysis results, the server provides improvement feedback to update the algorithm of the generative AI model.

[0885] The algorithm of the generative AI model is updated based on new patterns and techniques, and these are reflected in the next dialogue generation.

[0886] 2. The server deploys the system to a new version, providing users with the latest fraud techniques and countermeasures.

[0887] The system is automatically updated to reflect the latest fraud techniques and countermeasures.

[0888] In this way, the system not only allows users to experience real interactions with scammers, thereby raising their vigilance, but also allows users to analyze the generated data to understand scammers' behavioral patterns and new techniques, further improving the generative AI model.

[0889] Example 1

[0890] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0891] There is a lack of systems to raise awareness and vigilance against real-world fraud methods. While the risk of being victimized increases when personal information is carelessly disclosed, there are limited ways to learn effective prevention measures without experiencing actual fraud methods. In addition, there is a need for technology that can quickly respond to fraud methods that are constantly evolving.

[0892] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0893] In this invention, the server includes means for generating a profile of a fraudster character using a generative AI model, means for generating a dialogue scenario between the fraudster character and a user, means for presenting the generated dialogue scenario and receiving the user's responses, means for collecting and analyzing the received dialogue data, means for improving the generative AI model based on the analyzed data, means for receiving and authenticating authentication information, and means for updating the dialogue scenario based on the user's responses. This allows users to safely and effectively experience and learn realistic fraud techniques. Furthermore, the continuous improvement process using the generative AI model provides practical learning that always keeps up with the latest fraud techniques.

[0894] A "generative AI model" is an artificial intelligence technology that learns from a variety of data and generates new characters and dialogue scenarios.

[0895] A "profile" is information that details the characteristics and attributes of a particular character.

[0896] A "scammer character" is a virtual character that imitates fraudulent methods and reproduces the fraudulent tactics through dialogue with users.

[0897] A "dialogue scenario" is a script that comprises a series of dialogues that take place between the con man character and the user.

[0898] "User" means any person or entity that uses the System.

[0899] "Authentication information" refers to information used to verify a user's identity, and includes, for example, a username and password.

[0900] "Received dialogue data" is information regarding the dialogue that took place between the user and the impostor character.

[0901] "Collection and analysis" is the process of gathering interaction data and examining its content and trends.

[0902] "Improving a generative AI model" is the process of improving the performance and accuracy of an AI model based on collected and analyzed data.

[0903] "Updating" means to generate a new dialogue scenario based on the user's response and continue the dialogue scenario.

[0904] "Authentication" is the process of verifying that a user has valid access rights.

[0905] The system of the present invention comprises a server, a terminal, and a user. A specific embodiment will be described below.

[0906] Setting up the con man character

[0907] The server first generates a profile for the con man character using a generative AI model. Parameters such as age, gender, speech characteristics, and behavioral patterns are used to generate the profile. Specifically, the server inputs the following prompt sentence into the generative AI model:

[0908] "Generate a male in his 20s, polite-spoken, fraudulent technical support scammer character."

[0909] The generative AI model generates a profile of the imposter character based on this prompt sentence and saves the profile on the server.

[0910] User Login and Authentication

[0911] A user accesses the system using a terminal and enters authentication information (e.g., username and password) on the login screen. The terminal then sends this authentication information to the server.

[0912] The server checks the received authentication information against the database, and if authentication is successful, allows the user access. If invalid authentication information is entered, an error message is returned to the terminal.

[0913] Start of the Impostor Experience Session

[0914] After the user is successfully authenticated, they send a request to the server to start an impostor experience session. The server then sends a request to the generative AI model to generate a dialogue scenario with the impostor character.

[0915] The generative AI model generates a dialogue scenario for the con man character and returns the scenario to the server, which then sends the generated scenario to the device.

[0916] Recreating the fraud method

[0917] The terminal presents the dialogue scenario received from the server to the user in real time, and the user inputs a response according to the dialogue scenario, which is then sent to the server by the terminal.

[0918] The server receives the user's response and requests the generative AI model to continue the dialogue based on that response. The generative AI model generates a new dialogue and returns it to the server. The server then sends this new dialogue to the device. This process is repeated, and the user experiences a dialogue with the virtual impostor character.

[0919] Analysis and improvement of dialogue data

[0920] The server collects and stores all interaction data between the user and the virtual impostor character, which is stored in a database for analysis.

[0921] The server uses the collected dialogue data to provide feedback to the generative AI model, improving the behavior of the conman character and the algorithm for generating dialogue scenarios, making future dialogue scenarios more realistic and effective.

[0922] Specific examples

[0923] For example, to create a male con man character in his 20s, the server inputs conditions such as "male in his 20s, polite speech, fraudulent technical support" into the generative AI model. The generative AI model generates a specific con man character based on these conditions. When a user accesses the system from a terminal and starts a con man experience session, the following dialogue is generated:

[0924] Scammer: "Hello, this is tech support. Are you having trouble with your computer?"

[0925] User: "Yes, something seems to be wrong."

[0926] Scammer: "So, can you grant me remote access?"

[0927] Through these interactions, users can experience tech support scams and become more vigilant against them. The server collects and analyzes the interaction data and improves the generative AI model, enabling it to generate more accurate scammer characters and provide more accurate interaction scenarios.

[0928] Summary of prompt sentence examples

[0929] "Generate a male in his 20s, polite-spoken, fraudulent technical support scammer character."

[0930] "Generate a phishing scenario"

[0931] "Continue the dialogue of the imposter character based on the user's response."

[0932] This will enable users to learn specific countermeasures against various fraudulent methods and create a system that can prevent fraud damage before it occurs.

[0933] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0934] Step 1:

[0935] The server inputs a prompt sentence for generating a profile to the generative AI model.

[0936] Input: Character attribute information (e.g., age, gender, speech characteristics, behavior patterns)

[0937] Data processing: The generative AI model generates a character profile based on the input attribute information.

[0938] Output: Generated character profile

[0939] Specific operation: For example, a prompt such as "Male in his 20s, polite speaking, fraudulent technical support" is sent to the generation AI model, and a profile of the fraudster character is generated and stored on the server.

[0940] Step 2:

[0941] A user accesses the system using a terminal and enters authentication information (e.g., username and password) on a login screen.

[0942] Input: User authentication information (username, password)

[0943] Data processing: The terminal sends the entered authentication information to the server.

[0944] Output: Sending authentication information to the server

[0945] Specific operation: The user enters the username and password on the login screen and clicks the "Login" button.

[0946] Step 3:

[0947] The server checks the received authentication information against a database and allows the user access if authentication is successful.

[0948] Input: User authentication information sent from the device

[0949] Data processing: Check against the database to determine whether authentication was successful.

[0950] Output: Permissions or error messages

[0951] Specific operation: The server checks the authentication information stored in the database, and if it matches, notifies the terminal that authentication was successful.

[0952] Step 4:

[0953] After the user is successfully authenticated, the terminal sends a request to the server to initiate an impostor experience session.

[0954] Input: Session initiation request

[0955] Data processing: The server receives the request and begins processing.

[0956] Output: Session start confirmation message

[0957] Specific action: The user clicks the "Start Impostor Experience Session" button and submits the request.

[0958] Step 5:

[0959] The server sends a request to the generative AI model to generate an interaction scenario with the impostor character.

[0960] Input: Impostor Experience Session Start Request

[0961] Data processing: The generative AI model generates dialogue scenarios with the con man character.

[0962] Output: Generated dialogue scenario

[0963] Specific operation: The generative AI model generates a "tech support scam" scenario and returns it to the server.

[0964] Step 6:

[0965] The terminal presents the dialogue scenario received from the server to the user in real time.

[0966] Input: Interaction scenario sent from the server

[0967] Data processing: The scenario received by the terminal is displayed on the screen.

[0968] Output: Display of the interaction scenario to the user

[0969] What it does: The device displays the scammer character's line, "Hello, this is tech support. Are you having trouble with your computer?"

[0970] Step 7:

[0971] The user inputs a response according to the dialogue scenario, and the response is transmitted by the terminal to the server.

[0972] Input: User response (e.g. "Yes, something seems to be wrong.")

[0973] Data processing: The terminal sends the user's response to the server.

[0974] Output: Sending the user response to the server

[0975] Specific behavior: The user types a response and clicks the send button.

[0976] Step 8:

[0977] The server receives the user's response and, based on that, requests the generative AI model to continue the dialogue.

[0978] Input: User response

[0979] Data processing: Ask the generative AI model to continue the dialogue and generate a new scenario.

[0980] Output: Generated new dialogue scenario

[0981] Specific behavior: The generative AI model generates a new dialogue scenario and generates the reply, "So, may I grant you remote access?"

[0982] Step 9:

[0983] The server transmits the generated new dialogue scenario to the terminal.

[0984] Input: Generated new dialogue scenario

[0985] Data processing: Send the dialogue scenario to the terminal.

[0986] Output: Sending the dialogue scenario to the terminal

[0987] Specific operation: The server sends the newly generated dialogue scenario to the terminal, which then displays it to the user.

[0988] Step 10:

[0989] The server collects, stores, and analyzes all interaction data between the user and the virtual impostor character.

[0990] Input: Interaction data between the user and the impostor character

[0991] Data processing: Collection, storage, and analysis of interaction data

[0992] Output: Analysis results

[0993] What it does: The server stores the collected interaction data in a database and uses data analysis tools to extract patterns.

[0994] Step 11:

[0995] The server provides feedback to the generative AI model based on the analysis results, improving the AI ​​model.

[0996] Input: Analysis results

[0997] Data processing: Updating generative AI models

[0998] Output: An improved generative AI model

[0999] How it works: The generative AI model is refined based on newly collected data and analysis results, improving the quality of future dialogue scenarios.

[1000] (Application example 1)

[1001] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1002] Fraud methods continue to become more sophisticated, increasing the risk of many people becoming victims of fraud. Fraud methods via the Internet and telephone are particularly complex, making it difficult for many people to experience the risks in a realistic manner and raise their vigilance. Therefore, there is a need for a system that allows users to experience actual fraud methods and raise their vigilance.

[1003] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1004] In this invention, the server includes means for generating a profile of a fraudster character using a generative AI model, means for generating a dialogue scenario between the fraudster character and a user, and means for providing the generated dialogue scenario in a smartphone application, thereby enabling users to experience fraudulent tactics in a realistic manner through their smartphones and increase their vigilance.

[1005] A "generative AI model" is an artificial intelligence model that extracts specific patterns and information from data and generates natural language.

[1006] A "scammer character" is a profile of a virtual scammer created by a generative AI model, including age, gender, language, and behavioral patterns.

[1007] A "dialogue scenario" is a script that shows the flow of a pre-set conversation that takes place between a con man character and a user.

[1008] A "smartphone application" is a part of a program that runs on a smartphone and provides an interface for a user to interact with an impostor character.

[1009] "User response" refers to a reply or action that a user inputs or selects in a dialogue scenario.

[1010] "Dialogue data" is a record of the dialogue that took place between the user and the impostor character.

[1011] "Improving the generative AI model" refers to the process of analyzing collected interaction data and using the results to further improve the generative AI model.

[1012] A "fraudulent technique" is a specific method or means used by a fraudster to deceive a victim.

[1013] "Interactive experience to increase vigilance" refers to an educational dialogue that allows users to virtually experience fraudulent methods in order to increase their vigilance against real-life scams.

[1014] This invention relates to a system that uses a generative AI model to generate fraudster characters, allowing users to virtually experience fraud and raise their vigilance. The system consists of a server, a terminal (such as a smartphone), and a user.

[1015] Server Processing

[1016] The server uses a generative AI model to generate a profile of the fraudster character. To generate the profile, the user inputs conditions such as the fraudster's age, gender, speech characteristics, and behavioral patterns. This generates a specific fraudster character and stores it on the server.

[1017] The server then uses the generative AI model to generate a dialogue scenario for the user. This dialogue scenario is generated based on a specific fraudulent scheme (e.g., technical support fraud or phishing scam). The generated scenario is saved on the server and then sent to the user's device (smartphone).

[1018] Terminal (smartphone) processing

[1019] A user accesses the system using a smartphone and logs in. After logging in, the device sends the user's authentication information to the server, and authentication is performed. If authentication is successful, the server sends a dialogue scenario with the impostor character to the device and presents it to the user.

[1020] When the user initiates a dialogue, the device receives the user's response and sends it to the server. As the dialogue between the user and the impostor character progresses, the device continues to update the scenario in real time.

[1021] User Experience

[1022] The user experiences a virtual fraud through a smartphone application. For example, if the user selects the "tech support fraud" scenario, the following dialogue is generated:

[1023] Scammer Character: "Hello, tech support. Having trouble with your computer?"

[1024] User: "Yes, something seems to be wrong."

[1025] Impostor Character: "So, can you grant me remote access?"

[1026] Through such interactions, users can develop a sense of vigilance against real scams.

[1027] Data collection and analysis

[1028] Once the conversation ends, the server collects and stores the conversation data with the user. This data is fed back to the generative AI model, which analyzes the behavioral patterns of fraudsters and trends in fraudulent methods. Based on the analysis results, the generative AI model is improved. These improvements make future conversation scenarios more realistic and effective, improving the accuracy of replicating fraudulent methods.

[1029] Prompt Sentence Examples

[1030] Here's an example of a specific prompt for a generative AI model to generate a profile for a con man character:

[1031] "Male, 25 years old, polite speaking, fraudulent tech support scam. Generate a scenario to convince a user to grant remote access."

[1032] The present invention provides an effective means for users to experience virtual fraud through a smartphone application, thereby increasing their vigilance against real-life fraud.

[1033] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1034] Step 1:

[1035] The server inputs a prompt to the generative AI model to generate a profile of a con man character. This prompt includes the con man's age, gender, language, and behavioral patterns. Based on these conditions, the generative AI model generates a profile of a specific con man character and saves the profile on the server. The input data is the prompt, and the output data is the con man character's profile.

[1036] Step 2:

[1037] The server uses the generative AI model to generate a dialogue scenario between the conman character and the user. Specifically, the server inputs data on the conman character's profile and fraudulent methods into the generative AI model, and outputs a dialogue scenario. The generated dialogue scenario is saved on the server. The input data is the conman character's profile and fraudulent methods, and the output data is the dialogue scenario.

[1038] Step 3:

[1039] A user accesses the system using a smartphone terminal and logs in. The terminal sends the user's authentication information (username and password) to the server, which then authenticates it. If authentication is successful, the server sends the generated dialogue scenario to the terminal. The input data is the user's authentication information, and the output data is the authentication result and the dialogue scenario.

[1040] Step 4:

[1041] When the user selects to start the scenario, the smartphone device begins a dialogue with the imposter character. The user inputs a response to the imposter character's prompt, which is then sent from the device to the server. The server receives the user's response and updates the dialogue scenario as necessary. The input data is the user's response, and the output data is the updated dialogue scenario.

[1042] Step 5:

[1043] Once the interaction is complete, the server collects and stores the interaction data with the user. This interaction data includes the user's responses and the statements made by the imposter character. The server analyzes this data and feeds it back into the generative AI model. The input data is the interaction data, and the output data is the analysis results.

[1044] Step 6:

[1045] The server improves the generative AI model based on the analysis results. This makes the generation of future dialogue scenarios more realistic and effective, improving the accuracy of reproducing fraudulent methods. The input data is the analysis results, and the output data is the improved generative AI model.

[1046] Through the above processing steps, users can virtually experience fraud through a smartphone application, increasing their vigilance against real-life fraud.

[1047] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1048] This invention relates to a system that generates fraudster characters by combining a generative AI model and an emotion engine, and aims to improve users' understanding of fraudulent methods and their vigilance through dialogue between the user and the virtual fraudster character. This system consists of a server, a terminal, a user, and an emotion engine, and each component operates as follows:

[1049] Setting up the con man character

[1050] The server inputs conditions and parameters for generating a con artist character profile into the generative AI model. These conditions include the con artist's age, gender, speech characteristics, behavioral patterns, etc. The generative AI model uses these inputs to generate a specific con artist character profile, which is then stored on the server.

[1051] Authenticating the user and starting an interactive session

[1052] A user accesses the system using a terminal and logs in to begin an impostor experience session. The terminal sends the user's authentication information to the server, which verifies it. If authentication is successful, the server uses a generative AI model to generate an interaction scenario with the impostor character. This scenario is displayed on the user's terminal, and the user begins interacting with it.

[1053] Incorporating an emotion engine

[1054] The emotion engine analyzes video and audio data acquired from the user's camera and microphone to recognize the user's real-time emotional state. This emotional data is sent to the server and reflected in the responses of the con man character and the progression of the scenario.

[1055] Fraud reenactment and emotional feedback

[1056] The server uses the generative AI model to generate scenarios for recreating specific fraud schemes (e.g., phishing scams, technical support scams, etc.). The device presents the generated scenarios to the user in real time, and the emotion engine analyzes the user's emotional state. This allows the fraudster character to respond appropriately based on the user's emotional data. For example, if the user is feeling anxious, the fraudster character will use words that will further reassure them.

[1057] Collection and analysis of dialogue data

[1058] The server collects dialogue and emotional data between the user and the virtual con man character and stores it in a database. This data is fed back to the generative AI model and emotion engine, which analyzes trends in the con man's behavioral patterns and emotional responses. Based on the analysis results, both models are improved. These improvements make future dialogue scenarios more realistic and effective, improving the accuracy of reproducing fraudulent methods.

[1059] Specific examples

[1060] For example, let's create a male scammer character in his 20s on the theme of technical support fraud. The server inputs the following conditions into the generative AI model: "male in his 20s, polite speech, fraudulent technical support." Based on this, the generative AI model generates a specific scammer character. When a user accesses the system from a terminal and starts a scammer experience session, the following dialogue is generated:

[1061] Scammer: "Hello, this is tech support. Are you having trouble with your computer?"

[1062] User: "Yes, something seems to be wrong."

[1063] Scammer: "So, can you grant me remote access?"

[1064] During this dialogue, the emotion engine analyzes the user's facial expressions and voice to recognize their emotions in real time. If the emotion engine determines that the user is feeling anxious, the con man character will respond with a reassuring message, saying, "Remote operation is very safe and we can resolve the issue immediately." The server collects dialogue data and emotion data and feeds it back to the generative AI model and emotion engine to improve the accuracy of the system.

[1065] In this way, the system not only allows users to experience real interactions with scammers, thereby raising their vigilance, but also analyzes the generated data to understand scammers' behavioral patterns and new techniques, further improving the generative AI model and emotion engine.

[1066] The processing flow will be explained below.

[1067] Specific explanation of the program's processing steps

[1068] Step 1: Setting up your imposter character

[1069] 1. The server inputs the conditions and parameters for generating a profile of a con man character into the generative AI model.

[1070] The criteria include age, gender, language characteristics, and behavioral patterns.

[1071] 2. The generative AI model uses these conditions to generate a profile of a specific imposter character.

[1072] The profile includes the scammer's background, the fraudulent methods they use, and their luring techniques.

[1073] 3. The server stores the generated profile in a database.

[1074] Step 2: Authenticating the user and starting an interactive session

[1075] 1. A user accesses the system using a terminal and logs in.

[1076] The user enters authentication information (user name, password) in a dedicated interface.

[1077] 2. The terminal sends a login request to the server.

[1078] 3. The server checks the database and, if authentication is successful, allows the interactive session to begin.

[1079] If the authentication fails, an error message is returned to the terminal.

[1080] Step 3: Creating interactions with virtual characters

[1081] 1. The user operates the device and sends a request to the server to start a "scammer experience session."

[1082] The user clicks the "Start Trial Session" button within the interface.

[1083] 2. The server uses the generative AI model to generate dialogue scenarios with the impostor character.

[1084] The generative AI model creates initial dialogue scenarios and response patterns based on the profile of the conman character.

[1085] 3. The terminal displays the dialogue scenario to the user in real time and receives the user's responses.

[1086] Responses from the con man character are generated and displayed sequentially in response to user input.

[1087] Step 4: Replaying the scam and emotional feedback

[1088] 1. The server uses a generative AI model to generate scenarios to recreate specific fraud schemes.

[1089] The server instructs the AI ​​model on the type of fraudulent scheme it wants to replicate, such as "phishing scams" or "tech support scams."

[1090] 2. The generative AI model generates dialogue scenarios based on the specified fraudulent tactics.

[1091] The scenarios include specific steps on how the scammer will deceive the target.

[1092] 3. The emotion engine analyzes video and audio data obtained from the user's camera and microphone to recognize the user's emotional state.

[1093] Emotions are classified as, for example, "relief," "anxiety," and "doubt."

[1094] 4. The device displays the generated fraud scenario to the user, and the emotion engine analyzes the user's emotional state and transmits it to the server in real time.

[1095] The server then adjusts the imposter character's responses based on this: if the user is feeling anxious, the imposter character will respond in a reassuring way.

[1096] Step 5: Collect and analyze interaction data

[1097] 1. The server collects and stores in a database the conversation data and emotion data exchanged between the user and the impostor character.

[1098] All messages and emotional responses from each interaction session are logged.

[1099] 2. The server applies data analysis algorithms to analyze the collected interaction data and emotion data.

[1100] Identify trends in fraudster behavior patterns and user emotional responses.

[1101] Step 6: Improving the generative AI model and emotion engine

[1102] 1. Based on the analysis results, the server provides feedback to update the generative AI model algorithm and the emotion engine analysis.

[1103] The algorithm of the generative AI model is updated based on new patterns and techniques, and is reflected in the next dialogue generation. The emotion engine is also updated to improve the accuracy of the user's emotional response.

[1104] 2. The server deploys the system to a new version and provides users with an interactive experience that helps them understand and be aware of the latest fraud techniques.

[1105] The system is automatically updated to reflect the latest fraud techniques and countermeasures.

[1106] In this way, the system not only allows users to experience real-life interactions with scammers, thereby raising their vigilance, but also analyzes the generated data to understand scammers' behavioral patterns and new techniques, further improving the generative AI model and emotion engine.

[1107] Example 2

[1108] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1109] With the advancement of modern advanced communication and information technologies, fraud methods are becoming more sophisticated. In particular, fraud methods that skillfully exploit users' psychological state can cause serious damage. However, many users are not sufficiently vigilant against fraudulent methods and are easily deceived. Therefore, effective measures are needed to help users better understand fraudulent methods and increase their vigilance.

[1110] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1111] In this invention, the server includes means for generating a profile of a fraudster character using a generative AI model, means for generating a dialogue scenario between the fraudster character and the user, means for presenting the generated dialogue scenario to the user and receiving the user's response, means for acquiring real-time emotional data of the user using an emotion engine that recognizes the user's emotional state, means for collecting and analyzing the received dialogue data and emotional data, and means for improving the generative AI model based on the analyzed data. This allows the user to learn fraudulent tactics through dialogue with the fraudster character and to increase their vigilance against real-world fraud by receiving feedback on their emotional state in real time.

[1112] A "generative AI model" is a model that uses artificial intelligence technology to generate characters and dialogue scenarios tailored to specific purposes.

[1113] A "scammer character" is an artificially generated character that simulates fraudulent methods, and allows the user to learn fraudulent techniques and methods through dialogue with the user.

[1114] A "profile" is a set of information that includes characteristics such as the conman character's age, gender, speech patterns, and behavioral patterns.

[1115] A "dialogue scenario" specifically defines the flow and content of a series of dialogues that take place between a con man character and a user.

[1116] The "emotion engine" is a system that analyzes the user's facial expressions and voice through devices such as cameras and microphones, and recognizes their emotional state at that time in real time.

[1117] "Real-time" refers to reactions and analysis that occur immediately on the spot while the user is using the system.

[1118] "Emotion data" is information about the emotional state obtained from the user's facial expressions and voice analyzed by the emotion engine.

[1119] "Dialogue data" is information about the content of the conversation between the user and the impostor character, including the flow of the dialogue and specific wording.

[1120] "Analysis" refers to the process of using collected interaction and emotion data to evaluate patterns and trends in order to improve the performance of generative AI models and emotion engines.

[1121] This invention relates to a system that combines a generative AI model and an emotion engine to generate a fraudster character, and aims to improve understanding of fraudulent methods and vigilance through dialogue between the user and the virtual fraudster character. This system consists of a server, a terminal, a user, and an emotion engine, and each component operates as follows:

[1122] Setting up the con man character

[1123] The server inputs the conditions and parameters for generating a scammer character profile into the generative AI model. These conditions include the scammer's age, gender, speech characteristics, and behavioral patterns. For example, a prompt might read, "Please generate a character who is a man in his 20s who speaks politely and commits technical support scams." Based on these inputs, the generative AI model generates a profile for a specific scammer character, and the profile is stored on the server.

[1124] Authenticating the user and starting an interactive session

[1125] A user accesses the system using a terminal and logs in to begin an imposter experience session. The terminal sends the user's authentication information to the server, which verifies it. If authentication is successful, the server uses the generative AI model to generate an interaction scenario with the imposter character. For example, this can be achieved by sending the generative AI model a prompt such as "Please generate a tech support fraud scenario." This scenario is then displayed on the user's terminal, and the user begins interacting with it.

[1126] Incorporating an emotion engine

[1127] The emotion engine analyzes video and audio data acquired from the user's camera and microphone to recognize the user's real-time emotional state. This emotional data is sent to the server and reflected in the responses of the con man character and the progression of the scenario.

[1128] Fraud reenactment and emotional feedback

[1129] The server uses the generative AI model to generate scenarios for recreating specific fraud schemes (e.g., phishing scams, technical support scams, etc.). The device presents the generated scenarios to the user in real time, and the emotion engine analyzes the user's emotional state. This allows the fraudster character to respond appropriately based on the user's emotional data. For example, if the user is feeling anxious, the fraudster character will use words that will further reassure them.

[1130] Collection and analysis of dialogue data

[1131] The server collects dialogue and emotional data between the user and the virtual con man character and stores it in a database. This data is fed back to the generative AI model and emotion engine, which analyzes trends in the con man's behavioral patterns and emotional responses. Based on the analysis results, both models are improved. These improvements make future dialogue scenarios more realistic and effective, improving the accuracy of reproducing fraudulent methods.

[1132] Specific examples

[1133] For example, if you create a male scammer character in his 20s on the theme of technical support fraud, the server inputs the following conditions into the generative AI model: "male in his 20s, politely speaking, technical support fraud." Based on this, the generative AI model generates a specific scammer character.

[1134] When a user accesses the system from a terminal and starts an imposter experience session, the following dialogue is generated:

[1135] Scammer: "Hello, this is tech support. Are you having trouble with your computer?"

[1136] User: "Yes, something seems to be wrong."

[1137] Scammer: "So, can you grant me remote access?"

[1138] During this dialogue, the emotion engine analyzes the user's facial expressions and voice to recognize their emotions in real time. For example, if the emotion engine determines that the user is feeling anxious, the con artist character will respond with a reassuring "Remote operation is very safe, and we can resolve the issue immediately." The server collects dialogue and emotion data and feeds it back to the generative AI model and emotion engine to improve the system's accuracy. In this way, users not only experience realistic dialogue with con artists and become more vigilant, but also analyze the generated data to understand the con artists' behavioral patterns and new techniques, allowing them to make future dialogue scenarios more realistic and effective.

[1139] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1140] Step 1:

[1141] Setting up the con man character

[1142] The server inputs conditions and parameters for generating a profile of a con man character into the generative AI model. The input conditions include the con man's age, gender, speech characteristics, and behavioral patterns. The generative AI model then generates a profile of the con man character based on those conditions. The generated profile is stored in the server's database.

[1143] Specific operation: The server sends the prompt "Generate a character who is a man in his 20s and who will commit technical support scams in a polite tone" to the generation AI model and saves the profile of the resulting scammer character.

[1144] Input: Criteria for the conman character (age, gender, speech characteristics, behavioral patterns)

[1145] Output: Impostor character profile

[1146] Step 2:

[1147] Authenticating Users

[1148] A user accesses the system using a terminal and enters authentication information on the login screen. The terminal sends the authentication information to the server. The server compares the received authentication information with a database and determines whether the authentication is successful or not. If the authentication is successful, the server sends a session start notification to the user.

[1149] How it works: The user enters their username and password on the login screen, and the device sends this to the server, which then checks the information against the database and returns the results.

[1150] Input: User authentication information (username, password)

[1151] Output: Authentication result (success or failure)

[1152] Step 3:

[1153] Start of the Impostor Experience Session

[1154] The server generates a dialogue scenario with the impostor character using the generative AI model. The generated dialogue scenario is sent to the user's device and displayed. The user then begins a dialogue with the impostor character based on the scenario.

[1155] Specific operation: The server sends a prompt to the generative AI model saying, "Please generate a technical support fraud scenario," and presents the resulting scenario to the user's device.

[1156] Input: Prompt sentence (instruction to generate a dialogue scenario with the impostor character)

[1157] Output: Generated dialogue scenario

[1158] Step 4:

[1159] Incorporating an emotion engine

[1160] The device captures video and audio data from the user's camera and microphone. The emotion engine analyzes this data and recognizes the user's emotional state in real time. This emotional data is then sent to the server.

[1161] Specific operation: The device sends video and audio data to the emotion engine, which analyzes it to generate real-time emotion data and send it to the server.

[1162] Input: Video data, audio data

[1163] Output: User emotion data

[1164] Step 5:

[1165] Fraud reenactment and emotional feedback

[1166] The server uses the generative AI model to generate scenarios for recreating specific fraud schemes, which are then presented to the user, and the responses of the fraudster character are adjusted based on data from the emotion engine.

[1167] Specific operation: The server sends a prompt to the generative AI model to "generate a scenario that reproduces a fraudulent scheme," and presents the generated scenario to the user in real time. Based on the emotional data, the fraudster character's responses are appropriately adjusted.

[1168] Input: prompt sentence (instructions for reproducing the fraud scheme), user emotion data

[1169] Output: Adjusted impostor character response

[1170] Step 6:

[1171] Collection and analysis of dialogue data

[1172] The server collects and stores the conversation and emotion data between the user and the impostor character in a database. The collected data is fed back to the generative AI model and emotion engine, which then improves both models.

[1173] How it works: The server collects dialogue and emotion data in real time, stores it in a database, analyzes it, and uses it as feedback to improve the performance of the generative AI model and emotion engine.

[1174] Input: Dialogue data, emotion data

[1175] Output: Improved generative AI model, emotion engine

[1176] (Application example 2)

[1177] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1178] In modern society, fraudsters' tactics are becoming increasingly sophisticated, increasing the risk of ordinary users being scammed. Therefore, there is a need for effective training methods to help users understand actual fraudulent methods and increase their vigilance. However, current technology makes it difficult for users to interactively experience realistic fraud scenarios and receive feedback based on their emotional state while increasing their vigilance. To solve this problem, the development of an innovative system that combines a generative AI model and an emotion engine is desired.

[1179] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1180] In this invention, the server includes: means for generating a profile of a fraudster character using a generative AI model; means for generating a dialogue scenario between the fraudster character and a user; means for presenting the generated dialogue scenario to the user and receiving the user's response; means for collecting and analyzing the received dialogue data and the user's emotional data; means for improving the generative AI model and the emotion engine based on the analyzed data; and means including an emotion engine for adjusting the dialogue scenario based on the user's emotional data. This allows the user to experience fraudulent techniques in real time and receive appropriate feedback according to their emotional state, effectively increasing their vigilance against fraud.

[1181] A "generative AI model" is a form of generative artificial intelligence (AI) that is an algorithm that learns from large amounts of data and generates new profiles or text based on specific conditions and parameters.

[1182] A "scammer character" is a virtual scammer profile constructed using a generative AI model, and is a person with specific attributes such as age, gender, language, and behavioral patterns.

[1183] A "dialogue scenario" refers to the sequence or story of a dialogue that takes place between a con man character constructed by a generative AI model and the user.

[1184] An "emotion engine" is a general term for algorithms and software that analyze a user's facial expressions, voice, actions, etc. in real time and recognize their emotional state.

[1185] "Emotional Data" refers to information about a user's emotional state obtained and analyzed by the emotion engine.

[1186] "Dialogue data" refers to the record of the dialogue between the user and the virtual con man character and the text data thereof.

[1187] A "profile" refers to detailed information about a person generated based on specific conditions or parameters, including, for example, age, gender, speech, and behavioral patterns.

[1188] "Real-time" refers to processing and responses occurring almost immediately, without delay, and indicates a state in which dialogue with the user and emotional analysis are carried out immediately.

[1189] "Feedback" refers to the return of information that allows a generative AI model or emotion engine to change its response or improve the system based on the user's reaction or emotional state.

[1190] The present invention is a system that combines a generative AI model and an emotion engine to help users understand fraudulent tactics and raise their vigilance through dialogue with virtual fraudster characters. The system has the following specific configuration and operation procedures:

[1191] Server Roles

[1192] The server uses a generative AI model to generate a profile for the scammer character by inputting prompt statements containing criteria such as "male in his 20s, polite speech, technical support scam" into the model.

[1193] example:

[1194] text

[1195] Create a scammer character with the following attributes: 20-year-old male with polite speech, specializing in technical support scams.

[1196] Furthermore, the server uses the generative AI model to generate dialogue scenarios between the scammer character and the user, based on scamming techniques (e.g., phishing scams and tech support scams).

[1197] Device Role

[1198] The terminal used by the user is a device such as a smartphone or a head-mounted display. The terminal presents the dialogue scenario sent from the server to the user and receives the user's response.

[1199] The device also inputs video and audio data acquired from the user's camera and microphone into an emotion engine to recognize the user's real-time emotional state. The emotion engine uses software such as Affectiva. The emotion data is sent to the server and reflected in the responses of the imposter character and the progression of the scenario. For example, if the user shows signs of anxiety, the imposter character will respond reassuringly, saying, "Remote operation is very safe and we can solve the problem immediately."

[1200] User Roles

[1201] Users access the system via a terminal, log in, and begin a fraudster experience session. During the session, users interact with a virtual fraudster character and respond according to the progression of the dialogue scenario. Through this interaction, users can experience realistic fraudulent tactics and become more vigilant.

[1202] The server collects and analyzes user response and emotion data. The analyzed data is fed back to the generative AI model and emotion engine. This further improves the generative AI model and emotion engine, making future dialogue scenarios more realistic and effective.

[1203] These configurations and operational procedures allow users to experience fraudulent techniques in real time while receiving appropriate feedback based on their emotional state, which can increase their vigilance against fraud and help prevent them from becoming victims of fraud in real life.

[1204] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1205] Step 1:

[1206] The server uses a generative AI model to generate a profile for the fraudster character. In this process, the server sends a prompt (e.g., "Male in his 20s, polite speaking, tech support fraud") as input to the generative AI model. The generative AI model creates a profile for the fraudster character based on this prompt, and generates detailed information about the character (such as age, gender, speech, and behavioral patterns) as output, which is then stored on the server.

[1207] Step 2:

[1208] The server uses the generative AI model to generate a dialogue scenario between the scammer character and the user. In this scenario generation process, the server inputs data about fraudulent methods (e.g., specific methods of phishing scams) and sends it to the generative AI model. The dialogue scenario is generated based on this input data, and detailed text data of the scenario is obtained as output. The server then sends this scenario data to the user's device.

[1209] Step 3:

[1210] The terminal presents the dialogue scenario sent from the server to the user. At this stage, the user views the scenario through the terminal and begins dialogue with the virtual impostor character. The user's responses are input by the terminal and transmitted to the server.

[1211] Step 4:

[1212] The device inputs video and audio data acquired from the user's camera and microphone into the emotion engine to recognize the user's real-time emotional state. The emotion engine analyzes this input data and outputs the user's emotional state (e.g., anxiety, alertness, relaxation). This emotional data is sent to the server.

[1213] Step 5:

[1214] The server collects user response data and emotion data and provides feedback to the generative AI model and emotion engine. During this feedback process, the server performs data analysis based on the collected data and outputs parameters to improve the generative AI model and emotion engine. The generative AI model and emotion engine use this feedback to improve future dialogue scenarios and emotion recognition accuracy.

[1215] Step 6:

[1216] The server uses an improved generative AI model and emotion engine to adjust the impostor character's responses in real time based on the user's emotional state. For example, if the user is feeling anxious, the generative AI model generates a reassuring response such as "Remote operation is very safe and we can solve the problem immediately" and sends it to the device. The device then presents this response to the user and continues the dialogue.

[1217] Through these steps, the system allows users to experience fraudulent tactics in real time and provides feedback according to their emotional state, effectively raising their vigilance.

[1218] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1219] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1220] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1221] [Fourth embodiment]

[1222] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1223] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1224] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1225] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1226] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1227] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1228] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1229] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1230] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1231] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1232] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1233] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1234] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1235] This invention relates to a system that uses a generative AI model to generate fraudster characters, and aims to improve users' understanding of fraudulent methods and their vigilance through dialogue between the users and the virtual fraudster characters. This system consists of a server, a terminal, and a user, and each component operates as follows:

[1236] Setting up the con man character

[1237] The server inputs conditions and parameters for generating a con artist character profile into the generative AI model. These conditions include the con artist's age, gender, speech characteristics, behavioral patterns, etc. The generative AI model uses these inputs to generate a specific con artist character profile, which is then stored on the server.

[1238] Interaction generation with virtual characters

[1239] A user accesses the system using a terminal and logs in to begin an impostor experience session. The terminal sends the user's authentication information to the server, which verifies it. If authentication is successful, the server uses a generative AI model to generate an interaction scenario with the impostor character. This scenario is displayed on the user's terminal, and the user begins interacting with it.

[1240] Recreating the fraud method

[1241] The server uses the generative AI model to generate scenarios for recreating specific fraud methods (e.g., phishing scams, technical support scams, etc.). The device presents the generated scenarios to the user in real time and receives the user's responses. This allows the user to interact with the fraudster character and understand the fraud methods and patterns.

[1242] Analysis and improvement of dialogue data

[1243] The server collects and stores dialogue data between users and virtual con artists. This data is fed back to the generative AI model, which analyzes the behavioral patterns of con artists and trends in fraudulent methods. Based on the analysis results, the generative AI model is improved. These improvements make future dialogue scenarios more realistic and effective, improving the accuracy of replicating fraudulent methods.

[1244] Specific examples

[1245] For example, suppose you want to create a male con artist character in his 20s. The server inputs the following conditions into the generative AI model: "male in his 20s, polite speaking, fraudulent technical support." Based on this, the generative AI model generates a specific con artist character. When a user accesses the system from a terminal and starts a con artist experience session, the following dialogue is generated:

[1246] Scammer: "Hello, this is tech support. Are you having trouble with your computer?"

[1247] User: "Yes, something seems to be wrong."

[1248] Scammer: "So, can you grant me remote access?"

[1249] Through these interactions, users can experience the techniques used by technical support scammers and become more vigilant against them. The server collects and analyzes the interaction data and improves the AI ​​model, enabling it to generate more accurate scammer characters and provide more accurate interaction scenarios.

[1250] The processing flow will be explained below.

[1251] Specific explanation of the program's processing steps

[1252] Step 1: Setting up your imposter character

[1253] 1. The server inputs the conditions and parameters for generating a profile of a con man character into the generative AI model.

[1254] Specifically, information such as the scammer's age, gender, language characteristics, and behavioral patterns is set.

[1255] 2. A generative AI model uses these inputs to generate a profile of a specific imposter character.

[1256] The profile includes the scammer's background, the fraudulent methods they use, and how they lead you to the scam.

[1257] 3. The server stores the generated profile in a database.

[1258] Step 2: Authenticating the user and starting an interactive session

[1259] 1. The user accesses the system from a terminal and logs in.

[1260] The user enters authentication information (e.g., username and password) through a dedicated interface.

[1261] 2. The terminal sends a login request to the server.

[1262] 3. The server checks the database and, if authentication is successful, allows the interactive session to begin.

[1263] If the authentication fails, an error message is returned to the terminal.

[1264] Step 3: Creating interactions with virtual characters

[1265] 1. The user operates the device and sends a request to start an "impostor experience session."

[1266] The user clicks the "Start Trial Session" button within the interface.

[1267] 2. The server uses the generative AI model to generate dialogue scenarios with the impostor character.

[1268] The generative AI model creates initial dialogue scenarios and response patterns based on the profile of the conman character.

[1269] 3. The terminal displays the dialogue scenario to the user in real time and receives the user's responses.

[1270] Responses from the con man character are generated and displayed sequentially in response to user input.

[1271] Step 4: Recreate the scam

[1272] 1. The server uses a generative AI model to generate scenarios to recreate specific fraud schemes.

[1273] The server tells the AI ​​model what type of fraud scheme it wants to replicate, such as a "phishing scam" or a "tech support scam."

[1274] 2. The generative AI model generates dialogue scenarios based on the specified fraudulent tactics.

[1275] The scenario includes specific steps on how the scammer will deceive the target.

[1276] 3. The terminal displays the generated fraud scenario to the user and conducts a dialogue in real time.

[1277] By following the scenario, users can simulate the process of fraudulent schemes.

[1278] Step 5: Collect and analyze interaction data

[1279] 1. The server collects and stores in a database the interaction data between the user and the virtual impostor character.

[1280] All messages and responses that occur during each interactive session are collected as logs.

[1281] 2. The server applies a data analysis algorithm to analyze the collected interaction data.

[1282] Identify trends in fraudster behavior and tactics.

[1283] Step 6: Improving the generative AI model

[1284] 1. Based on the analysis results, the server provides improvement feedback to update the algorithm of the generative AI model.

[1285] The algorithm of the generative AI model is updated based on new patterns and techniques, and these are reflected in the next dialogue generation.

[1286] 2. The server deploys the system to a new version, providing users with the latest fraud techniques and countermeasures.

[1287] The system is automatically updated to reflect the latest fraud techniques and countermeasures.

[1288] In this way, the system not only allows users to experience real interactions with scammers, thereby raising their vigilance, but also allows users to analyze the generated data to understand scammers' behavioral patterns and new techniques, further improving the generative AI model.

[1289] Example 1

[1290] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1291] There is a lack of systems to raise awareness and vigilance against real-world fraud methods. While the risk of being victimized increases when personal information is carelessly disclosed, there are limited ways to learn effective prevention measures without experiencing actual fraud methods. In addition, there is a need for technology that can quickly respond to fraud methods that are constantly evolving.

[1292] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1293] In this invention, the server includes means for generating a profile of a fraudster character using a generative AI model, means for generating a dialogue scenario between the fraudster character and a user, means for presenting the generated dialogue scenario and receiving the user's responses, means for collecting and analyzing the received dialogue data, means for improving the generative AI model based on the analyzed data, means for receiving and authenticating authentication information, and means for updating the dialogue scenario based on the user's responses. This allows users to safely and effectively experience and learn realistic fraud techniques. Furthermore, the continuous improvement process using the generative AI model provides practical learning that always keeps up with the latest fraud techniques.

[1294] A "generative AI model" is an artificial intelligence technology that learns from a variety of data and generates new characters and dialogue scenarios.

[1295] A "profile" is information that details the characteristics and attributes of a particular character.

[1296] A "scammer character" is a virtual character that imitates fraudulent methods and reproduces the fraudulent tactics through dialogue with users.

[1297] A "dialogue scenario" is a script that comprises a series of dialogues that take place between the con man character and the user.

[1298] "User" means any person or entity that uses the System.

[1299] "Authentication information" refers to information used to verify a user's identity, and includes, for example, a username and password.

[1300] "Received dialogue data" is information regarding the dialogue that took place between the user and the impostor character.

[1301] "Collection and analysis" is the process of gathering interaction data and examining its content and trends.

[1302] "Improving a generative AI model" is the process of improving the performance and accuracy of an AI model based on collected and analyzed data.

[1303] "Updating" means to generate a new dialogue scenario based on the user's response and continue the dialogue scenario.

[1304] "Authentication" is the process of verifying that a user has valid access rights.

[1305] The system of the present invention comprises a server, a terminal, and a user. A specific embodiment will be described below.

[1306] Setting up the con man character

[1307] The server first generates a profile for the con man character using a generative AI model. Parameters such as age, gender, speech characteristics, and behavioral patterns are used to generate the profile. Specifically, the server inputs the following prompt sentence into the generative AI model:

[1308] "Generate a male in his 20s, polite-spoken, fraudulent technical support scammer character."

[1309] The generative AI model generates a profile of the imposter character based on this prompt sentence and saves the profile on the server.

[1310] User Login and Authentication

[1311] A user accesses the system using a terminal and enters authentication information (e.g., username and password) on the login screen. The terminal then sends this authentication information to the server.

[1312] The server checks the received authentication information against the database, and if authentication is successful, allows the user access. If invalid authentication information is entered, an error message is returned to the terminal.

[1313] Start of the Impostor Experience Session

[1314] After the user is successfully authenticated, they send a request to the server to start an impostor experience session. The server then sends a request to the generative AI model to generate a dialogue scenario with the impostor character.

[1315] The generative AI model generates a dialogue scenario for the con man character and returns the scenario to the server, which then sends the generated scenario to the device.

[1316] Recreating the fraud method

[1317] The terminal presents the dialogue scenario received from the server to the user in real time, and the user inputs a response according to the dialogue scenario, which is then sent to the server by the terminal.

[1318] The server receives the user's response and requests the generative AI model to continue the dialogue based on that response. The generative AI model generates a new dialogue and returns it to the server. The server then sends this new dialogue to the device. This process is repeated, and the user experiences a dialogue with the virtual impostor character.

[1319] Analysis and improvement of dialogue data

[1320] The server collects and stores all interaction data between the user and the virtual impostor character, which is stored in a database for analysis.

[1321] The server uses the collected dialogue data to provide feedback to the generative AI model, improving the behavior of the conman character and the algorithm for generating dialogue scenarios, making future dialogue scenarios more realistic and effective.

[1322] Specific examples

[1323] For example, to create a male con man character in his 20s, the server inputs conditions such as "male in his 20s, polite speech, fraudulent technical support" into the generative AI model. The generative AI model generates a specific con man character based on these conditions. When a user accesses the system from a terminal and starts a con man experience session, the following dialogue is generated:

[1324] Scammer: "Hello, this is tech support. Are you having trouble with your computer?"

[1325] User: "Yes, something seems to be wrong."

[1326] Scammer: "So, can you grant me remote access?"

[1327] Through these interactions, users can experience tech support scams and become more vigilant against them. The server collects and analyzes the interaction data and improves the generative AI model, enabling it to generate more accurate scammer characters and provide more accurate interaction scenarios.

[1328] Summary of prompt sentence examples

[1329] "Generate a male in his 20s, polite-spoken, fraudulent technical support scammer character."

[1330] "Generate a phishing scenario"

[1331] "Continue the dialogue of the imposter character based on the user's response."

[1332] This will enable users to learn specific countermeasures against various fraudulent methods and create a system that can prevent fraud damage before it occurs.

[1333] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1334] Step 1:

[1335] The server inputs a prompt sentence for generating a profile to the generative AI model.

[1336] Input: Character attribute information (e.g., age, gender, speech characteristics, behavior patterns)

[1337] Data processing: The generative AI model generates a character profile based on the input attribute information.

[1338] Output: Generated character profile

[1339] Specific operation: For example, a prompt such as "Male in his 20s, polite speaking, fraudulent technical support" is sent to the generation AI model, and a profile of the fraudster character is generated and stored on the server.

[1340] Step 2:

[1341] A user accesses the system using a terminal and enters authentication information (e.g., username and password) on a login screen.

[1342] Input: User authentication information (username, password)

[1343] Data processing: The terminal sends the entered authentication information to the server.

[1344] Output: Sending authentication information to the server

[1345] Specific operation: The user enters the username and password on the login screen and clicks the "Login" button.

[1346] Step 3:

[1347] The server checks the received authentication information against a database and allows the user access if authentication is successful.

[1348] Input: User authentication information sent from the device

[1349] Data processing: Check against the database to determine whether authentication was successful.

[1350] Output: Permissions or error messages

[1351] Specific operation: The server checks the authentication information stored in the database, and if it matches, notifies the terminal that authentication was successful.

[1352] Step 4:

[1353] After the user is successfully authenticated, the terminal sends a request to the server to initiate an impostor experience session.

[1354] Input: Session initiation request

[1355] Data processing: The server receives the request and begins processing.

[1356] Output: Session start confirmation message

[1357] Specific action: The user clicks the "Start Impostor Experience Session" button and submits the request.

[1358] Step 5:

[1359] The server sends a request to the generative AI model to generate an interaction scenario with the impostor character.

[1360] Input: Impostor Experience Session Start Request

[1361] Data processing: The generative AI model generates dialogue scenarios with the con man character.

[1362] Output: Generated dialogue scenario

[1363] Specific operation: The generative AI model generates a "tech support scam" scenario and returns it to the server.

[1364] Step 6:

[1365] The terminal presents the dialogue scenario received from the server to the user in real time.

[1366] Input: Interaction scenario sent from the server

[1367] Data processing: The scenario received by the terminal is displayed on the screen.

[1368] Output: Display of the interaction scenario to the user

[1369] What it does: The device displays the scammer character's line, "Hello, this is tech support. Are you having trouble with your computer?"

[1370] Step 7:

[1371] The user inputs a response according to the dialogue scenario, and the response is transmitted by the terminal to the server.

[1372] Input: User response (e.g. "Yes, something seems to be wrong.")

[1373] Data processing: The terminal sends the user's response to the server.

[1374] Output: Sending the user response to the server

[1375] Specific behavior: The user types a response and clicks the send button.

[1376] Step 8:

[1377] The server receives the user's response and, based on that, requests the generative AI model to continue the dialogue.

[1378] Input: User response

[1379] Data processing: Ask the generative AI model to continue the dialogue and generate a new scenario.

[1380] Output: Generated new dialogue scenario

[1381] Specific behavior: The generative AI model generates a new dialogue scenario and generates the reply, "So, may I grant you remote access?"

[1382] Step 9:

[1383] The server transmits the generated new dialogue scenario to the terminal.

[1384] Input: Generated new dialogue scenario

[1385] Data processing: Send the dialogue scenario to the terminal.

[1386] Output: Sending the dialogue scenario to the terminal

[1387] Specific operation: The server sends the newly generated dialogue scenario to the terminal, which then displays it to the user.

[1388] Step 10:

[1389] The server collects, stores, and analyzes all interaction data between the user and the virtual impostor character.

[1390] Input: Interaction data between the user and the impostor character

[1391] Data processing: Collection, storage, and analysis of interaction data

[1392] Output: Analysis results

[1393] What it does: The server stores the collected interaction data in a database and uses data analysis tools to extract patterns.

[1394] Step 11:

[1395] The server provides feedback to the generative AI model based on the analysis results, improving the AI ​​model.

[1396] Input: Analysis results

[1397] Data processing: Updating generative AI models

[1398] Output: An improved generative AI model

[1399] How it works: The generative AI model is refined based on newly collected data and analysis results, improving the quality of future dialogue scenarios.

[1400] (Application example 1)

[1401] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1402] Fraud methods continue to become more sophisticated, increasing the risk of many people becoming victims of fraud. Fraud methods via the Internet and telephone are particularly complex, making it difficult for many people to experience the risks in a realistic manner and raise their vigilance. Therefore, there is a need for a system that allows users to experience actual fraud methods and raise their vigilance.

[1403] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1404] In this invention, the server includes means for generating a profile of a fraudster character using a generative AI model, means for generating a dialogue scenario between the fraudster character and a user, and means for providing the generated dialogue scenario in a smartphone application, thereby enabling users to experience fraudulent tactics in a realistic manner through their smartphones and increase their vigilance.

[1405] A "generative AI model" is an artificial intelligence model that extracts specific patterns and information from data and generates natural language.

[1406] A "scammer character" is a profile of a virtual scammer created by a generative AI model, including age, gender, language, and behavioral patterns.

[1407] A "dialogue scenario" is a script that shows the flow of a pre-set conversation that takes place between a con man character and a user.

[1408] A "smartphone application" is a part of a program that runs on a smartphone and provides an interface for a user to interact with an impostor character.

[1409] "User response" refers to a reply or action that a user inputs or selects in a dialogue scenario.

[1410] "Dialogue data" is a record of the dialogue that took place between the user and the impostor character.

[1411] "Improving the generative AI model" refers to the process of analyzing collected interaction data and using the results to further improve the generative AI model.

[1412] A "fraudulent technique" is a specific method or means used by a fraudster to deceive a victim.

[1413] "Interactive experience to increase vigilance" refers to an educational dialogue that allows users to virtually experience fraudulent methods in order to increase their vigilance against real-life scams.

[1414] This invention relates to a system that uses a generative AI model to generate fraudster characters, allowing users to virtually experience fraud and raise their vigilance. The system consists of a server, a terminal (such as a smartphone), and a user.

[1415] Server Processing

[1416] The server uses a generative AI model to generate a profile of the fraudster character. To generate the profile, the user inputs conditions such as the fraudster's age, gender, speech characteristics, and behavioral patterns. This generates a specific fraudster character and stores it on the server.

[1417] The server then uses the generative AI model to generate a dialogue scenario for the user. This dialogue scenario is generated based on a specific fraudulent scheme (e.g., technical support fraud or phishing scam). The generated scenario is saved on the server and then sent to the user's device (smartphone).

[1418] Terminal (smartphone) processing

[1419] A user accesses the system using a smartphone and logs in. After logging in, the device sends the user's authentication information to the server, and authentication is performed. If authentication is successful, the server sends a dialogue scenario with the impostor character to the device and presents it to the user.

[1420] When the user initiates a dialogue, the device receives the user's response and sends it to the server. As the dialogue between the user and the impostor character progresses, the device continues to update the scenario in real time.

[1421] User Experience

[1422] The user experiences a virtual fraud through a smartphone application. For example, if the user selects the "tech support fraud" scenario, the following dialogue is generated:

[1423] Scammer Character: "Hello, tech support. Having trouble with your computer?"

[1424] User: "Yes, something seems to be wrong."

[1425] Impostor Character: "So, can you grant me remote access?"

[1426] Through such interactions, users can develop a sense of vigilance against real scams.

[1427] Data collection and analysis

[1428] Once the conversation ends, the server collects and stores the conversation data with the user. This data is fed back to the generative AI model, which analyzes the behavioral patterns of fraudsters and trends in fraudulent methods. Based on the analysis results, the generative AI model is improved. These improvements make future conversation scenarios more realistic and effective, improving the accuracy of replicating fraudulent methods.

[1429] Prompt Sentence Examples

[1430] Here's an example of a specific prompt for a generative AI model to generate a profile for a con man character:

[1431] "Male, 25 years old, polite speaking, fraudulent tech support scam. Generate a scenario to convince a user to grant remote access."

[1432] The present invention provides an effective means for users to experience virtual fraud through a smartphone application, thereby increasing their vigilance against real-life fraud.

[1433] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1434] Step 1:

[1435] The server inputs a prompt to the generative AI model to generate a profile of a con man character. This prompt includes the con man's age, gender, language, and behavioral patterns. Based on these conditions, the generative AI model generates a profile of a specific con man character and saves the profile on the server. The input data is the prompt, and the output data is the con man character's profile.

[1436] Step 2:

[1437] The server uses the generative AI model to generate a dialogue scenario between the conman character and the user. Specifically, the server inputs data on the conman character's profile and fraudulent methods into the generative AI model, and outputs a dialogue scenario. The generated dialogue scenario is saved on the server. The input data is the conman character's profile and fraudulent methods, and the output data is the dialogue scenario.

[1438] Step 3:

[1439] A user accesses the system using a smartphone terminal and logs in. The terminal sends the user's authentication information (username and password) to the server, which then authenticates it. If authentication is successful, the server sends the generated dialogue scenario to the terminal. The input data is the user's authentication information, and the output data is the authentication result and the dialogue scenario.

[1440] Step 4:

[1441] When the user selects to start the scenario, the smartphone device begins a dialogue with the imposter character. The user inputs a response to the imposter character's prompt, which is then sent from the device to the server. The server receives the user's response and updates the dialogue scenario as necessary. The input data is the user's response, and the output data is the updated dialogue scenario.

[1442] Step 5:

[1443] Once the interaction is complete, the server collects and stores the interaction data with the user. This interaction data includes the user's responses and the statements made by the imposter character. The server analyzes this data and feeds it back into the generative AI model. The input data is the interaction data, and the output data is the analysis results.

[1444] Step 6:

[1445] The server improves the generative AI model based on the analysis results. This makes the generation of future dialogue scenarios more realistic and effective, improving the accuracy of reproducing fraudulent methods. The input data is the analysis results, and the output data is the improved generative AI model.

[1446] Through the above processing steps, users can virtually experience fraud through a smartphone application, increasing their vigilance against real-life fraud.

[1447] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1448] This invention relates to a system that generates fraudster characters by combining a generative AI model and an emotion engine, and aims to improve users' understanding of fraudulent methods and their vigilance through dialogue between the user and the virtual fraudster character. This system consists of a server, a terminal, a user, and an emotion engine, and each component operates as follows:

[1449] Setting up the con man character

[1450] The server inputs conditions and parameters for generating a con artist character profile into the generative AI model. These conditions include the con artist's age, gender, speech characteristics, behavioral patterns, etc. The generative AI model uses these inputs to generate a specific con artist character profile, which is then stored on the server.

[1451] Authenticating the user and starting an interactive session

[1452] A user accesses the system using a terminal and logs in to begin an impostor experience session. The terminal sends the user's authentication information to the server, which verifies it. If authentication is successful, the server uses a generative AI model to generate an interaction scenario with the impostor character. This scenario is displayed on the user's terminal, and the user begins interacting with it.

[1453] Incorporating an emotion engine

[1454] The emotion engine analyzes video and audio data acquired from the user's camera and microphone to recognize the user's real-time emotional state. This emotional data is sent to the server and reflected in the responses of the con man character and the progression of the scenario.

[1455] Fraud reenactment and emotional feedback

[1456] The server uses the generative AI model to generate scenarios for recreating specific fraud schemes (e.g., phishing scams, technical support scams, etc.). The device presents the generated scenarios to the user in real time, and the emotion engine analyzes the user's emotional state. This allows the fraudster character to respond appropriately based on the user's emotional data. For example, if the user is feeling anxious, the fraudster character will use words that will further reassure them.

[1457] Collection and analysis of dialogue data

[1458] The server collects dialogue and emotional data between the user and the virtual con man character and stores it in a database. This data is fed back to the generative AI model and emotion engine, which analyzes trends in the con man's behavioral patterns and emotional responses. Based on the analysis results, both models are improved. These improvements make future dialogue scenarios more realistic and effective, improving the accuracy of reproducing fraudulent methods.

[1459] Specific examples

[1460] For example, let's create a male scammer character in his 20s on the theme of technical support fraud. The server inputs the following conditions into the generative AI model: "male in his 20s, polite speech, fraudulent technical support." Based on this, the generative AI model generates a specific scammer character. When a user accesses the system from a terminal and starts a scammer experience session, the following dialogue is generated:

[1461] Scammer: "Hello, this is tech support. Are you having trouble with your computer?"

[1462] User: "Yes, something seems to be wrong."

[1463] Scammer: "So, can you grant me remote access?"

[1464] During this dialogue, the emotion engine analyzes the user's facial expressions and voice to recognize their emotions in real time. If the emotion engine determines that the user is feeling anxious, the con man character will respond with a reassuring message, saying, "Remote operation is very safe and we can resolve the issue immediately." The server collects dialogue data and emotion data and feeds it back to the generative AI model and emotion engine to improve the accuracy of the system.

[1465] In this way, the system not only allows users to experience real interactions with scammers, thereby raising their vigilance, but also analyzes the generated data to understand scammers' behavioral patterns and new techniques, further improving the generative AI model and emotion engine.

[1466] The processing flow will be explained below.

[1467] Specific explanation of the program's processing steps

[1468] Step 1: Setting up your imposter character

[1469] 1. The server inputs the conditions and parameters for generating a profile of a con man character into the generative AI model.

[1470] The criteria include age, gender, language characteristics, and behavioral patterns.

[1471] 2. The generative AI model uses these conditions to generate a profile of a specific imposter character.

[1472] The profile includes the scammer's background, the fraudulent methods they use, and their luring techniques.

[1473] 3. The server stores the generated profile in a database.

[1474] Step 2: Authenticating the user and starting an interactive session

[1475] 1. A user accesses the system using a terminal and logs in.

[1476] The user enters authentication information (user name, password) in a dedicated interface.

[1477] 2. The terminal sends a login request to the server.

[1478] 3. The server checks the database and, if authentication is successful, allows the interactive session to begin.

[1479] If the authentication fails, an error message is returned to the terminal.

[1480] Step 3: Creating interactions with virtual characters

[1481] 1. The user operates the device and sends a request to the server to start a "scammer experience session."

[1482] The user clicks the "Start Trial Session" button within the interface.

[1483] 2. The server uses the generative AI model to generate dialogue scenarios with the impostor character.

[1484] The generative AI model creates initial dialogue scenarios and response patterns based on the profile of the conman character.

[1485] 3. The terminal displays the dialogue scenario to the user in real time and receives the user's responses.

[1486] Responses from the con man character are generated and displayed sequentially in response to user input.

[1487] Step 4: Replaying the scam and emotional feedback

[1488] 1. The server uses a generative AI model to generate scenarios to recreate specific fraud schemes.

[1489] The server instructs the AI ​​model on the type of fraudulent scheme it wants to replicate, such as "phishing scams" or "tech support scams."

[1490] 2. The generative AI model generates dialogue scenarios based on the specified fraudulent tactics.

[1491] The scenarios include specific steps on how the scammer will deceive the target.

[1492] 3. The emotion engine analyzes video and audio data obtained from the user's camera and microphone to recognize the user's emotional state.

[1493] Emotions are classified as, for example, "relief," "anxiety," and "doubt."

[1494] 4. The device displays the generated fraud scenario to the user, and the emotion engine analyzes the user's emotional state and transmits it to the server in real time.

[1495] The server then adjusts the imposter character's responses based on this: if the user is feeling anxious, the imposter character will respond in a reassuring way.

[1496] Step 5: Collect and analyze interaction data

[1497] 1. The server collects and stores in a database the conversation data and emotion data exchanged between the user and the impostor character.

[1498] All messages and emotional responses from each interaction session are logged.

[1499] 2. The server applies data analysis algorithms to analyze the collected interaction data and emotion data.

[1500] Identify trends in fraudster behavior patterns and user emotional responses.

[1501] Step 6: Improving the generative AI model and emotion engine

[1502] 1. Based on the analysis results, the server provides feedback to update the generative AI model algorithm and the emotion engine analysis.

[1503] The algorithm of the generative AI model is updated based on new patterns and techniques, and is reflected in the next dialogue generation. The emotion engine is also updated to improve the accuracy of the user's emotional response.

[1504] 2. The server deploys the system to a new version and provides users with an interactive experience that helps them understand and be aware of the latest fraud techniques.

[1505] The system is automatically updated to reflect the latest fraud techniques and countermeasures.

[1506] In this way, the system not only allows users to experience real-life interactions with scammers, thereby raising their vigilance, but also analyzes the generated data to understand scammers' behavioral patterns and new techniques, further improving the generative AI model and emotion engine.

[1507] Example 2

[1508] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1509] With the advancement of modern advanced communication and information technologies, fraud methods are becoming more sophisticated. In particular, fraud methods that skillfully exploit users' psychological state can cause serious damage. However, many users are not sufficiently vigilant against fraudulent methods and are easily deceived. Therefore, effective measures are needed to help users better understand fraudulent methods and increase their vigilance.

[1510] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1511] In this invention, the server includes means for generating a profile of a fraudster character using a generative AI model, means for generating a dialogue scenario between the fraudster character and the user, means for presenting the generated dialogue scenario to the user and receiving the user's response, means for acquiring real-time emotional data of the user using an emotion engine that recognizes the user's emotional state, means for collecting and analyzing the received dialogue data and emotional data, and means for improving the generative AI model based on the analyzed data. This allows the user to learn fraudulent tactics through dialogue with the fraudster character and to increase their vigilance against real-world fraud by receiving feedback on their emotional state in real time.

[1512] A "generative AI model" is a model that uses artificial intelligence technology to generate characters and dialogue scenarios tailored to specific purposes.

[1513] A "scammer character" is an artificially generated character that simulates fraudulent methods, and allows the user to learn fraudulent techniques and methods through dialogue with the user.

[1514] A "profile" is a set of information that includes characteristics such as the conman character's age, gender, speech patterns, and behavioral patterns.

[1515] A "dialogue scenario" specifically defines the flow and content of a series of dialogues that take place between a con man character and a user.

[1516] The "emotion engine" is a system that analyzes the user's facial expressions and voice through devices such as cameras and microphones, and recognizes their emotional state at that time in real time.

[1517] "Real-time" refers to reactions and analysis that occur immediately on the spot while the user is using the system.

[1518] "Emotion data" is information about the emotional state obtained from the user's facial expressions and voice analyzed by the emotion engine.

[1519] "Dialogue data" is information about the content of the conversation between the user and the impostor character, including the flow of the dialogue and specific wording.

[1520] "Analysis" refers to the process of using collected interaction and emotion data to evaluate patterns and trends in order to improve the performance of generative AI models and emotion engines.

[1521] This invention relates to a system that combines a generative AI model and an emotion engine to generate a fraudster character, and aims to improve understanding of fraudulent methods and vigilance through dialogue between the user and the virtual fraudster character. This system consists of a server, a terminal, a user, and an emotion engine, and each component operates as follows:

[1522] Setting up the con man character

[1523] The server inputs the conditions and parameters for generating a scammer character profile into the generative AI model. These conditions include the scammer's age, gender, speech characteristics, and behavioral patterns. For example, a prompt might read, "Please generate a character who is a man in his 20s who speaks politely and commits technical support scams." Based on these inputs, the generative AI model generates a profile for a specific scammer character, and the profile is stored on the server.

[1524] Authenticating the user and starting an interactive session

[1525] A user accesses the system using a terminal and logs in to begin an imposter experience session. The terminal sends the user's authentication information to the server, which verifies it. If authentication is successful, the server uses the generative AI model to generate an interaction scenario with the imposter character. For example, this can be achieved by sending the generative AI model a prompt such as "Please generate a tech support fraud scenario." This scenario is then displayed on the user's terminal, and the user begins interacting with it.

[1526] Incorporating an emotion engine

[1527] The emotion engine analyzes video and audio data acquired from the user's camera and microphone to recognize the user's real-time emotional state. This emotional data is sent to the server and reflected in the responses of the con man character and the progression of the scenario.

[1528] Fraud reenactment and emotional feedback

[1529] The server uses the generative AI model to generate scenarios for recreating specific fraud schemes (e.g., phishing scams, technical support scams, etc.). The device presents the generated scenarios to the user in real time, and the emotion engine analyzes the user's emotional state. This allows the fraudster character to respond appropriately based on the user's emotional data. For example, if the user is feeling anxious, the fraudster character will use words that will further reassure them.

[1530] Collection and analysis of dialogue data

[1531] The server collects dialogue and emotional data between the user and the virtual con man character and stores it in a database. This data is fed back to the generative AI model and emotion engine, which analyzes trends in the con man's behavioral patterns and emotional responses. Based on the analysis results, both models are improved. These improvements make future dialogue scenarios more realistic and effective, improving the accuracy of reproducing fraudulent methods.

[1532] Specific examples

[1533] For example, if you create a male scammer character in his 20s on the theme of technical support fraud, the server inputs the following conditions into the generative AI model: "male in his 20s, politely speaking, technical support fraud." Based on this, the generative AI model generates a specific scammer character.

[1534] When a user accesses the system from a terminal and starts an imposter experience session, the following dialogue is generated:

[1535] Scammer: "Hello, this is tech support. Are you having trouble with your computer?"

[1536] User: "Yes, something seems to be wrong."

[1537] Scammer: "So, can you grant me remote access?"

[1538] During this dialogue, the emotion engine analyzes the user's facial expressions and voice to recognize their emotions in real time. For example, if the emotion engine determines that the user is feeling anxious, the con artist character will respond with a reassuring "Remote operation is very safe, and we can resolve the issue immediately." The server collects dialogue and emotion data and feeds it back to the generative AI model and emotion engine to improve the system's accuracy. In this way, users not only experience realistic dialogue with con artists and become more vigilant, but also analyze the generated data to understand the con artists' behavioral patterns and new techniques, allowing them to make future dialogue scenarios more realistic and effective.

[1539] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1540] Step 1:

[1541] Setting up the con man character

[1542] The server inputs conditions and parameters for generating a profile of a con man character into the generative AI model. The input conditions include the con man's age, gender, speech characteristics, and behavioral patterns. The generative AI model then generates a profile of the con man character based on those conditions. The generated profile is stored in the server's database.

[1543] Specific operation: The server sends the prompt "Generate a character who is a man in his 20s and who will commit technical support scams in a polite tone" to the generation AI model and saves the profile of the resulting scammer character.

[1544] Input: Criteria for the conman character (age, gender, speech characteristics, behavioral patterns)

[1545] Output: Impostor character profile

[1546] Step 2:

[1547] Authenticating Users

[1548] A user accesses the system using a terminal and enters authentication information on the login screen. The terminal sends the authentication information to the server. The server compares the received authentication information with a database and determines whether the authentication is successful or not. If the authentication is successful, the server sends a session start notification to the user.

[1549] How it works: The user enters their username and password on the login screen, and the device sends this to the server, which then checks the information against the database and returns the results.

[1550] Input: User authentication information (username, password)

[1551] Output: Authentication result (success or failure)

[1552] Step 3:

[1553] Start of the Impostor Experience Session

[1554] The server generates a dialogue scenario with the impostor character using the generative AI model. The generated dialogue scenario is sent to the user's device and displayed. The user then begins a dialogue with the impostor character based on the scenario.

[1555] Specific operation: The server sends a prompt to the generative AI model saying, "Please generate a technical support fraud scenario," and presents the resulting scenario to the user's device.

[1556] Input: Prompt sentence (instruction to generate a dialogue scenario with the impostor character)

[1557] Output: Generated dialogue scenario

[1558] Step 4:

[1559] Incorporating an emotion engine

[1560] The device captures video and audio data from the user's camera and microphone. The emotion engine analyzes this data and recognizes the user's emotional state in real time. This emotional data is then sent to the server.

[1561] Specific operation: The device sends video and audio data to the emotion engine, which analyzes it to generate real-time emotion data and send it to the server.

[1562] Input: Video data, audio data

[1563] Output: User emotion data

[1564] Step 5:

[1565] Fraud reenactment and emotional feedback

[1566] The server uses the generative AI model to generate scenarios for recreating specific fraud schemes, which are then presented to the user, and the responses of the fraudster character are adjusted based on data from the emotion engine.

[1567] Specific operation: The server sends a prompt to the generative AI model to "generate a scenario that reproduces a fraudulent scheme," and presents the generated scenario to the user in real time. Based on the emotional data, the fraudster character's responses are appropriately adjusted.

[1568] Input: prompt sentence (instructions for reproducing the fraud scheme), user emotion data

[1569] Output: Adjusted impostor character response

[1570] Step 6:

[1571] Collection and analysis of dialogue data

[1572] The server collects and stores the conversation and emotion data between the user and the impostor character in a database. The collected data is fed back to the generative AI model and emotion engine, which then improves both models.

[1573] How it works: The server collects dialogue and emotion data in real time, stores it in a database, analyzes it, and uses it as feedback to improve the performance of the generative AI model and emotion engine.

[1574] Input: Dialogue data, emotion data

[1575] Output: Improved generative AI model, emotion engine

[1576] (Application example 2)

[1577] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1578] In modern society, fraudsters' tactics are becoming increasingly sophisticated, increasing the risk of ordinary users being scammed. Therefore, there is a need for effective training methods to help users understand actual fraudulent methods and increase their vigilance. However, current technology makes it difficult for users to interactively experience realistic fraud scenarios and receive feedback based on their emotional state while increasing their vigilance. To solve this problem, the development of an innovative system that combines a generative AI model and an emotion engine is desired.

[1579] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1580] In this invention, the server includes: means for generating a profile of a fraudster character using a generative AI model; means for generating a dialogue scenario between the fraudster character and a user; means for presenting the generated dialogue scenario to the user and receiving the user's response; means for collecting and analyzing the received dialogue data and the user's emotional data; means for improving the generative AI model and the emotion engine based on the analyzed data; and means including an emotion engine for adjusting the dialogue scenario based on the user's emotional data. This allows the user to experience fraudulent techniques in real time and receive appropriate feedback according to their emotional state, effectively increasing their vigilance against fraud.

[1581] A "generative AI model" is a form of generative artificial intelligence (AI) that is an algorithm that learns from large amounts of data and generates new profiles or text based on specific conditions and parameters.

[1582] A "scammer character" is a virtual scammer profile constructed using a generative AI model, and is a person with specific attributes such as age, gender, language, and behavioral patterns.

[1583] A "dialogue scenario" refers to the sequence or story of a dialogue that takes place between a con man character constructed by a generative AI model and the user.

[1584] An "emotion engine" is a general term for algorithms and software that analyze a user's facial expressions, voice, actions, etc. in real time and recognize their emotional state.

[1585] "Emotional Data" refers to information about a user's emotional state obtained and analyzed by the emotion engine.

[1586] "Dialogue data" refers to the record of the dialogue between the user and the virtual con man character and the text data thereof.

[1587] A "profile" refers to detailed information about a person generated based on specific conditions or parameters, including, for example, age, gender, speech, and behavioral patterns.

[1588] "Real-time" refers to processing and responses occurring almost immediately, without delay, and indicates a state in which dialogue with the user and emotional analysis are carried out immediately.

[1589] "Feedback" refers to the return of information that allows a generative AI model or emotion engine to change its response or improve the system based on the user's reaction or emotional state.

[1590] The present invention is a system that combines a generative AI model and an emotion engine to help users understand fraudulent tactics and raise their vigilance through dialogue with virtual fraudster characters. The system has the following specific configuration and operation procedures:

[1591] Server Roles

[1592] The server uses a generative AI model to generate a profile for the scammer character by inputting prompt statements containing criteria such as "male in his 20s, polite speech, technical support scam" into the model.

[1593] example:

[1594] text

[1595] Create a scammer character with the following attributes: 20-year-old male with polite speech, specializing in technical support scams.

[1596] Furthermore, the server uses the generative AI model to generate dialogue scenarios between the scammer character and the user, based on scamming techniques (e.g., phishing scams and tech support scams).

[1597] Device Role

[1598] The terminal used by the user is a device such as a smartphone or a head-mounted display. The terminal presents the dialogue scenario sent from the server to the user and receives the user's response.

[1599] The device also inputs video and audio data acquired from the user's camera and microphone into an emotion engine to recognize the user's real-time emotional state. The emotion engine uses software such as Affectiva. The emotion data is sent to the server and reflected in the responses of the imposter character and the progression of the scenario. For example, if the user shows signs of anxiety, the imposter character will respond reassuringly, saying, "Remote operation is very safe and we can solve the problem immediately."

[1600] User Roles

[1601] Users access the system via a terminal, log in, and begin a fraudster experience session. During the session, users interact with a virtual fraudster character and respond according to the progression of the dialogue scenario. Through this interaction, users can experience realistic fraudulent tactics and become more vigilant.

[1602] The server collects and analyzes user response and emotion data. The analyzed data is fed back to the generative AI model and emotion engine. This further improves the generative AI model and emotion engine, making future dialogue scenarios more realistic and effective.

[1603] These configurations and operational procedures allow users to experience fraudulent techniques in real time while receiving appropriate feedback based on their emotional state, which can increase their vigilance against fraud and help prevent them from becoming victims of fraud in real life.

[1604] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1605] Step 1:

[1606] The server uses a generative AI model to generate a profile for the fraudster character. In this process, the server sends a prompt (e.g., "Male in his 20s, polite speaking, tech support fraud") as input to the generative AI model. The generative AI model creates a profile for the fraudster character based on this prompt, and generates detailed information about the character (such as age, gender, speech, and behavioral patterns) as output, which is then stored on the server.

[1607] Step 2:

[1608] The server uses the generative AI model to generate a dialogue scenario between the scammer character and the user. In this scenario generation process, the server inputs data about fraudulent methods (e.g., specific methods of phishing scams) and sends it to the generative AI model. The dialogue scenario is generated based on this input data, and detailed text data of the scenario is obtained as output. The server then sends this scenario data to the user's device.

[1609] Step 3:

[1610] The terminal presents the dialogue scenario sent from the server to the user. At this stage, the user views the scenario through the terminal and begins dialogue with the virtual impostor character. The user's responses are input by the terminal and transmitted to the server.

[1611] Step 4:

[1612] The device inputs video and audio data acquired from the user's camera and microphone into the emotion engine to recognize the user's real-time emotional state. The emotion engine analyzes this input data and outputs the user's emotional state (e.g., anxiety, alertness, relaxation). This emotional data is sent to the server.

[1613] Step 5:

[1614] The server collects user response data and emotion data and provides feedback to the generative AI model and emotion engine. During this feedback process, the server performs data analysis based on the collected data and outputs parameters to improve the generative AI model and emotion engine. The generative AI model and emotion engine use this feedback to improve future dialogue scenarios and emotion recognition accuracy.

[1615] Step 6:

[1616] The server uses an improved generative AI model and emotion engine to adjust the impostor character's responses in real time based on the user's emotional state. For example, if the user is feeling anxious, the generative AI model generates a reassuring response such as "Remote operation is very safe and we can solve the problem immediately" and sends it to the device. The device then presents this response to the user and continues the dialogue.

[1617] Through these steps, the system allows users to experience fraudulent tactics in real time and provides feedback according to their emotional state, effectively raising their vigilance.

[1618] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1619] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1620] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1621] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1622] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1623] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1624] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1625] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1626] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1627] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1628] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1629] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1630] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1631] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[1632] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1633] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1634] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1635] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1636] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1637] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1638] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1639] The following is further disclosed regarding the above embodiment.

[1640] (Claim 1)

[1641] a means for generating a profile of a fraudster character using a generative AI model;

[1642] means for generating a dialogue scenario between the imposter character and a user;

[1643] means for presenting the generated dialogue scenario to a user and receiving a response from the user;

[1644] means for collecting and analyzing the received interaction data;

[1645] means for improving a generative AI model based on the analyzed data;

[1646] A system including:

[1647] (Claim 2)

[1648] 10. The system of claim 1, further comprising: means for generating a dialogue scenario based on a particular fraud technique to reproduce the fraud technique.

[1649] (Claim 3)

[1650] The system of claim 1 , further comprising means for progressing an interaction scenario with the generated impostor character in real time to provide a user with an interaction experience that heightens vigilance.

[1651] "Example 1"

[1652] (Claim 1)

[1653] a means for generating a profile of a fraudster character using a generative AI model;

[1654] means for generating a dialogue scenario between the impostor character and a user;

[1655] means for presenting the generated dialogue scenario to a user and receiving a response from the user;

[1656] means for collecting and analyzing the received interaction data;

[1657] means for improving a generative AI model based on the analyzed data;

[1658] means for receiving and authenticating authentication information;

[1659] means for updating the dialogue scenario based on a user's response;

[1660] A system including:

[1661] (Claim 2)

[1662] 10. The system of claim 1, further comprising: means for generating an interaction scenario based on a particular fraud technique to replicate the fraud technique.

[1663] (Claim 3)

[1664] The system according to claim 1, further comprising means for progressing an interaction scenario with the generated impostor character in real time to provide a user with an interaction experience that heightens vigilance.

[1665] "Application Example 1"

[1666] (Claim 1)

[1667] a means for generating a profile of a fraudster character using a generative AI model;

[1668] means for generating a dialogue scenario between the imposter character and a user;

[1669] means for presenting the generated dialogue scenario to a user and receiving a response from the user;

[1670] means for collecting and analyzing the received interaction data;

[1671] means for improving a generative AI model based on the analyzed data;

[1672] means for providing the dialogue scenario in a smartphone application;

[1673] means for providing an interactive experience using said smartphone application to increase fraud alertness;

[1674] A system including:

[1675] (Claim 2)

[1676] 10. The system of claim 1, further comprising: means for generating a dialogue scenario based on a particular fraud technique to reproduce the fraud technique.

[1677] (Claim 3)

[1678] The system of claim 1 , further comprising means for progressing an interaction scenario with the generated impostor character in real time to provide a user with an interaction experience that heightens vigilance.

[1679] "Example 2: Combining Emotion Engines"

[1680] (Claim 1)

[1681] a means for generating a profile of a fraudster character using a generative AI model;

[1682] means for generating a dialogue scenario between the imposter character and a user;

[1683] means for presenting the generated dialogue scenario to a user and receiving a response from the user;

[1684] means for acquiring real-time emotional data of a user using an emotion engine that recognizes the emotional state of the user;

[1685] means for collecting and analyzing the received dialogue data and emotion data;

[1686] means for improving a generative AI model based on the analyzed data;

[1687] A system including:

[1688] (Claim 2)

[1689] 10. The system of claim 1, further comprising means for generating a dialogue scenario based on a particular fraud technique and for progressing a dialogue with the generated fraudster character in real time.

[1690] (Claim 3)

[1691] 10. The system of claim 1, further comprising means for providing an interactive experience in which the imposter character provides appropriate responses based on the user's emotional data to heighten the user's vigilance.

[1692] "Application example 2 when combining emotion engines"

[1693] (Claim 1)

[1694] a means for generating a profile of a fraudster character using a generative AI model;

[1695] means for generating a dialogue scenario between the imposter character and a user;

[1696] means for presenting the generated dialogue scenario to a user and receiving a response from the user;

[1697] means for collecting and analyzing the received dialogue data and user emotion data;

[1698] means for improving the generative AI model and emotion engine based on the analyzed data;

[1699] means including an emotion engine for adjusting a dialogue scenario based on emotion data of the user;

[1700] A system including:

[1701] (Claim 2)

[1702] 10. The system of claim 1, further comprising: means for generating a dialogue scenario based on a particular fraud technique to reproduce the fraud technique.

[1703] (Claim 3)

[1704] The system of claim 1 , further comprising means for progressing an interaction scenario with the generated impostor character in real time to provide a user with an interaction experience that heightens vigilance. [Explanation of symbols]

[1705] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. a means for generating a profile of a fraudster character using a generative AI model; means for generating a dialogue scenario between the imposter character and a user; means for presenting the generated dialogue scenario to a user and receiving a response from the user; means for collecting and analyzing the received interaction data; means for improving a generative AI model based on the analyzed data; A system including:

2. The system of claim 1 , further comprising: means for generating a dialogue scenario based on a specific fraud technique to reproduce the fraud technique.

3. The system according to claim 1 , further comprising means for progressing an interaction scenario with the generated impostor character in real time, and providing a user with an interaction experience that heightens vigilance.

Images