System

Abstract

To provide a system.SOLUTION: A system includes means for: detecting potentially fraudulent information among information communicated to a user; suggesting a response to the potentially fraudulent information; and analyzing an emotional state of the user and providing advice to reduce at least one of stress and anxiety of the user.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] It is difficult to respond appropriately and ensure safety against the sophisticated speech techniques and documents of fraudsters who access your system via telephone or email. [Means for solving the problem]

[0005] The present invention detects anything unusual in the speech or documents of fraudsters and proposes responses to uncover them. It also provides a system that communicates and responds to fraudsters and connects only those deemed safe to users, enabling users to communicate safely. [Brief explanation of the drawings]

[0006] [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. 2 is a sequence diagram showing a flow of processing in the data processing system according to the first embodiment of the first form example. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1 of Embodiment 1. [Figure 13] FIG. 10 is a sequence diagram showing a processing flow of a data processing system in a second embodiment of the second form example. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 of Embodiment Example 2. [Figure 15] FIG. 10 is a sequence diagram showing the flow of processing in a data processing system according to a third embodiment of the third embodiment. [Figure 16] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 3 of Embodiment 3. [Figure 17] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in the first embodiment of the first form example when an emotion engine is combined. [Figure 18] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1 of Form Example 1 when an emotion engine is combined. [Figure 19] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in the second embodiment of the second form example when an emotion engine is combined. [Figure 20] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 of Form Example 2 when an emotion engine is combined. [Figure 21] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in the third embodiment of the third form example when an emotion engine is combined. [Figure 22] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 3 of Form Example 3 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0007] 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.

[0008] First, the terms used in the following description will be explained.

[0009] 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, the 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), an APU (Accelerated Processing Unit), or a TPU (TENSOR PROCESSING UNIT (registered trademark)).

[0010] 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.

[0011] 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.

[0012] 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.

[0013] 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."

[0014] [First embodiment]

[0015] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0016] 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.

[0017] 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).

[0018] 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.

[0019] 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.

[0020] 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.

[0021] 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.

[0022] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0023] 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.

[0024] 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.

[0025] 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.

[0026] Next, the specific processing by the specific processing unit 290 of the data processing device 12 will be described.

[0027] "Example 1"

[0028] One embodiment of the present invention provides a means for detecting unusual content in the speech techniques and documents of sophisticated fraudsters who access a user via telephone, email, etc. This means analyzes the speech techniques and documents of the fraudster and determines whether they contain fraudulent elements. For example, it can detect situations such as when a fraudster makes inappropriate requests to a user or when a fraudster provides information that misleads a user.

[0029] "Example 2"

[0030] Also provided is a means for suggesting responses to uncover the content of a fraudster. This means generates appropriate responses to the fraudster's speech or document content and provides them to the user. For example, if a fraudster makes an inappropriate request to the user, it is possible to suggest a response that ignores the request. Also, if a fraudster provides information that misleads the user, it is possible to suggest a response that denies that information.

[0031] "Example 3"

[0032] Furthermore, a means for communicating and responding to fraudsters is also provided. This means communicates with fraudsters on behalf of the user and reports the results to the user. For example, if a fraudster makes an inappropriate request of the user, the user can respond to the fraudster by ignoring the request. Also, if a fraudster provides information that misleads the user, the user can respond to the fraudster by denying that information.

[0033] "Example 4"

[0034] Finally, a method is provided to connect users only with content that is determined to be safe. This method provides content to users only if the scammer's speech or document content is determined to be safe as a result of interactions with the scammer. For example, if the scammer provides appropriate information to the user, that information can be provided to the user. Also, if the scammer makes inappropriate requests of the user or provides information that is misleading, such information will not be provided to the user.

[0035] The processing flow of each embodiment will be described below.

[0036] "Example 1"

[0037] Step 1: A method for detecting unusual content is activated in the speech techniques and documents of sophisticated fraudsters who access the victim via telephone, email, etc. This method analyzes the speech techniques and contents of documents of fraudsters and determines whether they contain any fraudulent elements.

[0038] Step 2: Generate appropriate responses to the scammer's speech and document content and provide them to the user. In this step, if the scammer makes an inappropriate request to the user, a response that ignores the request is suggested.

[0039] Step 3: Interact with the fraudster on behalf of the user and report the results to the user. In this step, if the fraudster makes an inappropriate request to the user, the system responds to the fraudster by ignoring the request.

[0040] Step 4: If the fraudster's language and document content are deemed safe after interaction with the fraudster, the content is provided to the user. In this step, if the fraudster has provided the user with appropriate information, the information is provided to the user.

[0041] "Example 2"

[0042] Step 1: A method for detecting unusual content is activated in the speech techniques and documents of sophisticated fraudsters who access the victim via telephone, email, etc. This method analyzes the speech techniques and contents of documents of fraudsters and determines whether they contain any fraudulent elements.

[0043] Step 2: Generate appropriate responses to the scammer's speech and document content and provide them to the user. In this step, if the scammer makes an inappropriate request to the user, a response that ignores the request is suggested.

[0044] Step 3: Interact with the fraudster on behalf of the user and report the results to the user. In this step, if the fraudster makes an inappropriate request to the user, the system responds to the fraudster by ignoring the request.

[0045] Step 4: If the fraudster's language and document content are deemed safe after interaction with the fraudster, the content is provided to the user. In this step, if the fraudster has provided the user with appropriate information, the information is provided to the user.

[0046] "Example 3"

[0047] Step 1: A method for detecting unusual content is activated in the speech techniques and documents of sophisticated fraudsters who access the victim via telephone, email, etc. This method analyzes the speech techniques and contents of documents of fraudsters and determines whether they contain any fraudulent elements.

[0048] Step 2: Generate appropriate responses to the scammer's speech and document content and provide them to the user. In this step, if the scammer makes an inappropriate request to the user, a response that ignores the request is suggested.

[0049] Step 3: Interact with the fraudster on behalf of the user and report the results to the user. In this step, if the fraudster makes an inappropriate request to the user, the system responds to the fraudster by ignoring the request.

[0050] Step 4: If the fraudster's language and document content are deemed safe after interaction with the fraudster, the content is provided to the user. In this step, if the fraudster has provided the user with appropriate information, the information is provided to the user.

[0051] "Example 4"

[0052] Step 1: A method for detecting unusual content is activated in the speech techniques and documents of sophisticated fraudsters who access the victim via telephone, email, etc. This method analyzes the speech techniques and contents of documents of fraudsters and determines whether they contain any fraudulent elements.

[0053] Step 2: Generate appropriate responses to the scammer's speech and document content and provide them to the user. In this step, if the scammer makes an inappropriate request to the user, a response that ignores the request is suggested.

[0054] Step 3: Interact with the fraudster on behalf of the user and report the results to the user. In this step, if the fraudster makes an inappropriate request to the user, the system responds to the fraudster by ignoring the request.

[0055] Step 4: If the fraudster's language and document content are deemed safe after interaction with the fraudster, the content is provided to the user. In this step, if the fraudster has provided the user with appropriate information, the information is provided to the user.

[0056] Example 1

[0057] Next, a description will be given of Example 1 of Form 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."

[0058] In modern society, there has been an increase in the number of sophisticated fraud methods carried out via telephone and email. This has increased the risk of many users becoming victims of fraud. Conventional methods have made it difficult to detect and respond to fraud, requiring users to rely on their own judgment to spot fraud. For this reason, there is a need for a system that automates fraud detection and response and protects users.

[0059] 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.

[0060] In this invention, the server includes a means for a user to input a message, a means for the terminal to send the message to the server, a means for the server to analyze the message, a means for the server to determine fraudulent elements using a generative AI model, a means for the server to generate an evaluation result, a means for the server to send the evaluation result to the terminal, and a means for the terminal to display the evaluation result to the user. This automates fraud detection and response, making it possible to reduce the risk of users becoming victims of fraud.

[0061] "Telephone and email" refers to a method of conveying information to users using communication means.

[0062] A "sophisticated fraudster" is someone who uses sophisticated verbal and written techniques to deceive users.

[0063] "Methods for detecting unusual content" are methods for finding unusual or unnatural elements in a message.

[0064] A "means for suggesting response content to detect fraudulent content" is a method for generating appropriate responses to potentially fraudulent messages.

[0065] "Methods of interacting with and responding to scammers" refers to methods for managing communications with scammers and responding appropriately.

[0066] "Means of connecting users only with messages that are deemed safe" is a method of providing users with only messages that are deemed to pose no risk of fraud.

[0067] "Means for users to input messages" refers to the method by which users input messages they receive into the system.

[0068] "Means for a terminal to send a message to a server" refers to a method by which a user's device sends an input message to a server.

[0069] The "means by which the server analyzes the message" refers to a method by which the server analyzes the message received using a natural language processing tool.

[0070] The "means by which the server uses the generative AI model to determine fraudulent elements" is the method by which the server uses the generative AI model to evaluate the fraudulent elements of a message.

[0071] The "means by which the server generates the evaluation result" refers to the method by which the server creates the evaluation result for the message based on the output of the generative AI model.

[0072] The "means by which the server transmits the evaluation results to the terminal" refers to a method by which the server transmits the evaluation results generated by the server to the user's device.

[0073] The "means by which the terminal displays the evaluation results to the user" refers to a method by which the user's device displays the evaluation results it has received on its screen.

[0074] This invention is a system that protects users by detecting any unusual content in the speech techniques and documents of sophisticated fraudsters who access them via telephone, email, etc. This system is realized using a server, terminals, a generative AI model, and a natural language processing tool.

[0075] Hardware and software used

[0076] Hardware: Servers, devices (PCs, smartphones, etc.)

[0077] Software: Generative AI models (e.g., GPT-4®), natural language processing tools (e.g., spaCy, NLTK)

[0078] Specific operation of the system

[0079] 1. The user types a message

[0080] Users input the contents of phone calls and emails they receive into the system. For example, if a user receives an email that they suspect is fraudulent, they can copy the contents of the email and paste it into the system's input field.

[0081] 2. The device sends a message to the server

[0082] The device (user's PC or smartphone) sends the entered message to the server. Specifically, it sends the message data to the server using an HTTP request.

[0083] 3. The server parses the message

[0084] The server analyzes the received message using natural language processing tools (e.g., spaCy or NLTK). Specifically, it analyzes the grammatical structure of the message and extracts important keywords and phrases.

[0085] 4. The server uses a generative AI model to determine fraudulent elements

[0086] The server uses a generative AI model (e.g., GPT-4) to determine whether the message content contains fraudulent elements. Specifically, the extracted keywords and phrases are incorporated into a prompt sentence and input into the generative AI model. For example, the prompt sentence "Does this message contain fraudulent elements?" is input into the generative AI model.

[0087] 5. The server generates the evaluation result

[0088] The server evaluates whether the message is suspicious based on the output of the generative AI model. Specifically, it analyzes the output of the generative AI model and generates an evaluation result such as "Possible fraud" or "Not likely fraud."

[0089] 6. The server sends the evaluation results to the device.

[0090] The server sends the generated evaluation results to the terminal. Specifically, it sends the evaluation results to the terminal using an HTTP response.

[0091] 7. The device displays the evaluation results to the user.

[0092] The device then displays the evaluation results to the user. Specifically, it displays a warning message on the user's PC or smartphone screen, such as "This message may be fraudulent."

[0093] Specific examples

[0094] Email example

[0095] Input message: "Congratulations! You are the winner of 1 million yen. Click this link for more details."

[0096] Prompt Sentence Examples

[0097] Prompt: "Does this message contain any elements of deception?"

[0098] Example output

[0099] Generative AI model output: "This message is fraudulent, especially since it asks you to click on a link."

[0100] Evaluation results

[0101] Server rating: "Possibly a scam"

[0102] Display results

[0103] Device displays: "This message may be fraudulent. Do not click on any links."

[0104] In this way, users can detect potentially fraudulent messages in advance and prevent damage before it occurs. This system automates fraud detection and response, reducing the risk of users becoming victims of fraud.

[0105] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0106] Step 1:

[0107] The user enters a message.

[0108] Input: The contents of phone calls and emails received by the user.

[0109] What happens: A user copies and pastes a suspected fraudulent message into an input field on the system.

[0110] Output: The input message data.

[0111] Step 2:

[0112] The device sends a message to the server.

[0113] Input: Message data entered by the user.

[0114] Specific operation: The device (user's PC or smartphone) sends message data to the server using an HTTP request.

[0115] Output: The message data sent to the server.

[0116] Step 3:

[0117] The server parses the message.

[0118] Input: The message data sent to the server.

[0119] What it does: The server uses natural language processing tools (e.g., spaCy or NLTK) to analyze the grammatical structure of the message and extract important keywords and phrases.

[0120] Output: Extracted keywords and phrases.

[0121] Step 4:

[0122] The server uses a generative AI model to determine fraudulent elements.

[0123] Input: Extracted keywords or phrases.

[0124] How it works: The server inputs a prompt sentence like "Does this message contain any elements of fraud?" into a generative AI model (e.g., GPT-4) and obtains the model's output.

[0125] Output: The output of the generative AI model (a rating of whether or not there is fraud).

[0126] Step 5:

[0127] The server generates the evaluation results.

[0128] Input: The output of a generative AI model.

[0129] Specific operation: The server analyzes the output of the generative AI model and generates an evaluation result such as "Possible fraud" or "Not likely to be fraudulent."

[0130] Output: Evaluation result.

[0131] Step 6:

[0132] The server transmits the evaluation results to the terminal.

[0133] Input: Evaluation result.

[0134] Specific operation: The server sends the evaluation results to the terminal using an HTTP response.

[0135] Output: Evaluation results sent to the device.

[0136] Step 7:

[0137] The terminal displays the evaluation results to the user.

[0138] Input: The evaluation result sent to the device.

[0139] Specific behavior: The device displays a warning message on the user's PC or smartphone screen, such as "This message may be fraudulent."

[0140] Output: The evaluation results displayed to the user.

[0141] (Application example 1)

[0142] Next, a description will be given of Application Example 1 of Embodiment 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."

[0143] In recent years, sophisticated frauds using telephone and email have been increasing, increasing the risk of users becoming victims of fraud. In particular, it is difficult for users to detect frauds on their own when fraudsters use sophisticated speaking and writing techniques. Therefore, effective measures are needed to reduce the risk of fraud and ensure the safety of users.

[0144] 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.

[0145] In this invention, the server includes means for detecting anything unusual in the speech techniques or documents of sophisticated fraudsters who access the server by phone or email, means for suggesting responses to uncover the fraudster's content, means for communicating and responding to the fraudster, means for connecting only those determined to be safe to the user, means for analyzing voice data and detecting fraudulent elements, means for analyzing text data and detecting fraudulent elements, and means for issuing a warning if there is a possibility of fraud. This reduces the risk that users will be involved in fraud and ensures safe communication.

[0146] "Means for detecting unusual content" refers to a function that analyzes information received via telephone, email, etc., and automatically detects fraudulent elements or unnatural points.

[0147] The "means for suggesting response content to detect the content of a fraudster" is a function that generates appropriate responses to the fraudster's speech or writing and suggests them to the user.

[0148] "Means of interacting and responding to scammers" refers to functions for managing communication with scammers and taking appropriate action.

[0149] "Means of connecting users with only those things that are determined to be safe" is a function that provides users with only information and communications that are determined to pose no risk of fraud.

[0150] "Means for analyzing voice data and detecting fraudulent elements" refers to a function that analyzes voice data such as telephone conversations and detects elements that may be fraudulent.

[0151] "Means for analyzing text data and detecting fraudulent elements" refers to a function that analyzes text data such as emails and messages and detects elements that may be fraudulent.

[0152] "Means for issuing warnings in the event of possible fraud" is a function that issues a warning to the user when a risk of fraud is detected.

[0153] A system for implementing this invention includes means for detecting anything unusual in the speech techniques or documents of sophisticated fraudsters who access the system by phone or email, means for suggesting responses to uncover the fraudster's content, means for communicating and responding to fraudsters, means for connecting users only with those determined to be safe, means for analyzing voice data and detecting fraudulent elements, means for analyzing text data and detecting fraudulent elements, and means for issuing a warning if there is a possibility of fraud.

[0154] The server uses a speech recognition library to analyze the audio data and converts the audio file into text. The converted text is then analyzed by a generative AI model (e.g., a model using the transformers library) to determine whether it contains fraudulent elements. If fraudulent elements are detected, the server issues a warning to the user.

[0155] In addition, text data (e.g., emails and messages) is also analyzed by the generative AI model to determine whether it contains fraudulent elements. If fraudulent elements are detected, the server issues a warning to the user.

[0156] For example, if a user receives a message saying, "Please tell me your account information," the system will analyze the message and issue a warning if it determines there is a possibility of fraud. An example of a prompt sentence to input into the generative AI model is, "Please tell me your account information."

[0157] This system allows users to reduce the risk of being involved in fraud and ensure safe communication.

[0158] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0159] Step 1:

[0160] The server obtains the voice data or text data received from the user.

[0161] Input: Audio or text data

[0162] Output: Acquired audio data or text data

[0163] Step 2:

[0164] The server uses a speech recognition library to convert the voice data into text.

[0165] Input: Audio data

[0166] Output: Converted text data

[0167] Specific operation: Reads an audio file, performs speech recognition, and outputs it as text.

[0168] Step 3:

[0169] The server inputs the converted text data or directly received text data into the generative AI model.

[0170] Input: Text data

[0171] Output: Analysis results from the generative AI model

[0172] What it does: Text data is fed into a generative AI model (e.g., a model using the transformers library) and analyzed for deceptive elements.

[0173] Step 4:

[0174] The server determines whether the data contains any fraudulent elements based on the analysis results of the generative AI model.

[0175] Input: Analysis results of the generative AI model

[0176] Output: Judgment result on whether there is any fraudulent element

[0177] What it does: Evaluate the analysis results and flag any fraudulent elements detected.

[0178] Step 5:

[0179] The server will warn the user if any fraudulent elements are detected.

[0180] Input: Judgment result regarding the presence or absence of fraudulent elements

[0181] Output: Warning message

[0182] What it does: Sends a warning message to the user if there is a possibility of fraud.

[0183] Step 6:

[0184] If the server does not detect any fraudulent elements, it determines that the transaction is safe and proceeds with normal processing for the user.

[0185] Input: Judgment result regarding the presence or absence of fraudulent elements

[0186] Output: Normal processing result

[0187] What it does: Provide users with information or communications that are determined to be free of fraud risks.

[0188] Example 2

[0189] Next, a description will be given of Example 2 of Form 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."

[0190] In modern society, sophisticated fraud methods via telephone and email are on the rise. This increases the risk that many users will be deceived by the scammers' verbal tactics and documents, and lose their personal information and money. With conventional methods, it is difficult to properly analyze the scammers' messages and respond quickly and accurately, making it difficult to ensure the safety of users.

[0191] 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.

[0192] In this invention, the server includes means for receiving messages from fraudsters, means for transmitting the received messages to the server, means for inputting the messages received by the server into a generative AI model, means for the generative AI model to analyze the fraudster's message and generate an appropriate response content, means for transmitting the generated response content to the user's terminal, means for the user to check the suggested response content and reply to the fraudster as necessary, and means for connecting only those messages that are determined to be safe to the user. This makes it possible to respond quickly and accurately to fraudster's messages and ensure the safety of users.

[0193] A "fraudster" is someone who attempts to fraudulently obtain money or information by deceiving others.

[0194] A "message" is a communication sent by a fraudster to a user, and may be in the form of text, audio, images, etc.

[0195] A "server" is a computer system that processes and stores data over a network.

[0196] A "terminal" is a device that is directly operated by a user, and includes smartphones, tablets, personal computers, etc.

[0197] A "generative AI model" is an algorithm that uses artificial intelligence technology to generate and analyze text, such as a natural language processing model.

[0198] A "prompt" is an instruction entered into a generative AI model, which serves as a trigger for a specific response or analysis.

[0199] The "response content" is the message generated by the generative AI model as a result of analysis, and includes an appropriate response to the scammer.

[0200] "Analysis" is the process by which a generative AI model understands the input message and extracts meaning.

[0201] "Things that have been determined to be safe" refer to communication content and parties that have been determined to pose no risk of fraud.

[0202] "User" means an individual or organization that uses the system to respond to messages from scammers.

[0203] MODE FOR CARRYING OUT THE INVENTION

[0204] The present invention provides a system that generates appropriate responses to messages from fraudsters and helps users respond safely. A specific embodiment of this system is described below.

[0205] Hardware and software used

[0206] Hardware: Servers, terminals

[0207] Software: Generative AI models (e.g., GPT-4)

[0208] System Overview

[0209] The system receives messages from fraudsters, analyzes the messages, and generates appropriate responses. The generated responses are provided to the user, who can then use them to respond to the fraudster.

[0210] Specific actions

[0211] 1. The user receives a message from a scammer

[0212] A user receives a message from a scammer. For example, the scammer asks for your bank details.

[0213] 2. The device sends the received message to the server

[0214] The device sends the received message to the server. Specifically, the device application sends the message to the server via the API.

[0215] 3. The server inputs the received message into the generative AI model.

[0216] The server inputs the received message into the generative AI model. Specifically, the server passes the message to the generative AI model as a prompt. For example, the prompt might be, "I received a message from a scammer saying, 'Please tell me your bank account information.' Please suggest an appropriate response to this message."

[0217] 4. A generative AI model analyzes the scammer's message and generates an appropriate response.

[0218] The generative AI model analyzes the prompt and generates an appropriate response to the scammer's message, for example, "Sorry, but we can't give out any personal information. Let us know if there's anything else we can help you with."

[0219] 5. The server sends the generated response to the user's device.

[0220] The server sends the response received from the generated AI model to the user's device. Specifically, the server sends the response to the device via API.

[0221] 6. The user reviews the suggested responses and responds to the scammer if necessary

[0222] The user checks the suggested responses on their device. The user then uses the suggested responses to reply to the scammer. For example, the user might reply, "Sorry, but I can't give out any personal information."

[0223] Specific examples

[0224] For example, if a user receives the message "Please tell me your bank account information," the following prompt sentence is input into the generative AI model:

[0225] Example prompt sentence:

[0226] I received a message from a scammer asking for my bank details. Can you suggest an appropriate response?

[0227] Example response of the generative AI model:

[0228] "Sorry, I can't give out any personal information. Let me know if there's anything else I can help you with."

[0229] In this way, when a user receives a message from a scammer, the system uses a generative AI model to suggest an appropriate response, helping the user respond safely.

[0230] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0231] Step 1:

[0232] The user receives a message from the scammer.

[0233] Input: A message from the scammer (e.g., "Please tell me your bank details")

[0234] Output: Received message

[0235] What happens: The user's device receives and displays a message from the scammer.

[0236] Step 2:

[0237] The terminal sends the received message to the server.

[0238] Input: Received message

[0239] Output: Message sent to the server

[0240] Specific operation: The device application sends the received message to the server via API.

[0241] Step 3:

[0242] The server inputs the received message into the generative AI model.

[0243] Input: The message sent to the server

[0244] Output: The prompt input to the generative AI model

[0245] Specific operation: The server passes the received message to the generative AI model as a prompt. For example, the prompt might be, "I received a message from a scammer saying, 'Please tell me your bank account information.' Please suggest an appropriate response to this message."

[0246] Step 4:

[0247] A generative AI model analyzes the scammer's message and generates an appropriate response.

[0248] Input: prompt statement

[0249] Output: The generated response

[0250] How it works: The generative AI model analyzes the prompt and generates an appropriate response to the scammer's message. For example, the generative AI model might generate a response like, "Sorry, we can't give out any personal information. Let us know if there's anything else we can help you with."

[0251] Step 5:

[0252] The server sends the generated response to the user's terminal.

[0253] Input: Generated response content

[0254] Output: The response sent to the user's device

[0255] Specific operation: The server sends the response received from the generated AI model to the user's device via API.

[0256] Step 6:

[0257] The user reviews the suggested responses and responds to the scammer if desired.

[0258] Input: The response sent to the user's device

[0259] Output: Reply to the scammer

[0260] What happens: The user reviews the suggested responses on their device and responds to the scammer if necessary. For example, the user might reply, "Sorry, I can't give out my personal information."

[0261] (Application example 2)

[0262] Next, a description will be given of Application Example 2 of Form 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."

[0263] In recent years, sophisticated fraudulent activities using telephone and email have been increasing, increasing the risk of users being deceived by the scammers' verbal techniques and documents. There is a need for a method that allows users to respond appropriately to such fraudulent activities. There is also a need for a system that can detect signs of fraud in real time and provide appropriate responses.

[0264] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for detecting anything unusual in the speech techniques or documents of a clever fraudster who accesses the server by phone, email, etc., means for suggesting responses to uncover the fraudster's content, means for communicating and responding to the fraudster, means for connecting only those determined to be safe to the user, means for recording audio and converting the audio to text, means for determining the possibility of fraud using a generative AI model and generating an appropriate response, and means for displaying the generated response to the user. This allows users to reduce the risk of fraud and communicate safely.

[0265] "Telephone, email, etc." refers to means of communication such as voice calls and email.

[0266] A "sophisticated fraudster" is someone who uses sophisticated verbal and written techniques to deceive users.

[0267] "Means for detecting unusual content" refers to technology that automatically identifies unusual content, including indicators of fraud.

[0268] "Means for suggesting response content to detect fraudsters' content" refers to technology for generating appropriate responses to fraudsters' requests and information and providing them to users.

[0269] "Methods of interacting with and responding to fraudsters" refers to techniques used to manage communications with fraudsters and respond appropriately.

[0270] "Means of connecting users only with communications that are determined to be safe" refers to technology that provides users with only communications that are determined to be void of fraud risks.

[0271] "Means for recording audio and converting audio to text" refers to technology for recording audio data and converting that audio into text information.

[0272] "Means for using generative AI models to determine likelihood of fraud and generate appropriate responses" refers to technologies for using artificial intelligence models to assess likelihood of fraud and generate appropriate responses.

[0273] The term "means for displaying the generated response to the user" refers to a technique for visually presenting the generated response content to the user.

[0274] The system for implementing this invention has the function of detecting anything unusual in the speech techniques or documents of clever fraudsters who access the system by telephone or email, proposing responses to uncover the fraudster's content, communicating and responding to the fraudster, and connecting only those deemed safe to the user.

[0275] System configuration

[0276] The system consists of the following main components:

[0277] 1. Audio Recording Device: Use your smartphone's microphone to record audio.

[0278] 2. Speech Recognition Software: Converts recorded speech into text using a speech recognition library.

[0279] 3. Generative AI model: Uses OpenAI® APIs to analyze text data and determine the likelihood of fraud.

[0280] 4. Response Generator: Uses generative AI models to generate appropriate responses in cases of potential fraud.

[0281] 5. Display device: The smartphone display is used to display the generated response to the user.

[0282] Processing flow

[0283] 1. Voice recording and transcription: When a user answers a phone call, a voice recorder records the conversation, which is then converted into text by speech recognition software.

[0284] 2. Fraud Detection: A generative AI model analyzes text data to determine the likelihood of fraud, for example, using the following prompt:

[0285] Determine whether the following text is a scam and suggest an appropriate response:

[0286] Please tell us your bank account details

[0287] 3. Generate and display a response: If a fraudulent request is detected, the generative AI model generates an appropriate response and displays it to the user via a display device. For example, a response such as "Please ignore this request. It is extremely dangerous to give your bank account information to others" could be generated.

[0288] Specific examples

[0289] If a user receives the message "Please tell us your bank account information," the system operates as follows:

[0290] 1. A voice recorder records your conversation, which speech recognition software converts into text: "Please tell me your bank account details."

[0291] 2. A generative AI model analyzes this text and determines that it is likely to be fraudulent.

[0292] 3. The response generating device generates a response saying, "Please ignore this request. It is extremely dangerous to give your bank account information to others," and displays it to the user on the display device.

[0293] In this way, users can communicate safely and with less risk of fraud.

[0294] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0295] Step 1:

[0296] When the user answers the phone, the device's voice recorder records the conversation. The input is the conversation between the user and the fraudster, and the output is the recorded voice data, which is used in subsequent processing steps.

[0297] Step 2:

[0298] The device's speech recognition software (speech recognition library) converts the recorded voice data into text. The input is the recorded voice data, and the output is text data. This text data is used for fraud detection.

[0299] Step 3:

[0300] The server's generative AI model (OpenAI API) analyzes the text data and determines the likelihood of fraud. The input is the text data, and the output is the evaluation result regarding the likelihood of fraud. Specifically, the analysis is performed using the following prompt sentence:

[0301] Determine whether the following text is a scam and suggest an appropriate response:

[0302] Please tell us your bank account details

[0303] Step 4:

[0304] The server's response generator generates an appropriate response based on the evaluation results of the generative AI model. The input is the evaluation result regarding the possibility of fraud, and the output is the generated response text. For example, a response might be generated that reads, "Please ignore this request. It is very dangerous to give your bank account information to others."

[0305] Step 5:

[0306] The display device of the terminal displays the generated response text to the user. The input is the generated response text, and the output is the response information visually provided to the user. This allows users to communicate safely, reducing the risk of fraud.

[0307] Example 3

[0308] Next, a third embodiment of the third embodiment will be described. 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."

[0309] There is a problem that fraudsters try to deceive users by using clever speech techniques and documents over the phone, email, etc. Conventional systems have difficulty detecting fraudsters' fraudulent activities and responding appropriately, which increases the risk of users becoming victims of fraud. In addition, there is a lack of means to automate interactions with fraudsters and provide users with only safe information. This requires an effective way to prevent fraudsters' fraudulent activities while ensuring the safety of users.

[0310] The specific processing by the specific processing unit 290 of the data processing device 12 in the third embodiment is realized by the following means.

[0311] In this invention, the server includes means for receiving messages from fraudsters, means for analyzing the received messages, means for generating responses to the fraudsters using a generative AI model, means for sending the generated responses to the fraudsters, means for saving the results of interactions with the fraudsters, means for extracting safe information from the results of the interactions, and means for providing the safe information to users. This makes it possible to automatically detect fraudulent activities by fraudsters and respond appropriately. Furthermore, by providing only safe information to users, the safety of users can be ensured and fraud victims can be prevented.

[0312] A "fraudster" is someone who uses fraudulent means to deceive others and obtain money or information illegally.

[0313] A "message" is a communication sent by a scammer to a user, and may be in the form of text, audio, images, etc.

[0314] "Analysis" is the process of analyzing the content of received messages to determine their intent and whether or not they contain fraudulent activity.

[0315] A "generative AI model" is a system that uses artificial intelligence technology to generate and analyze text and speech, such as a natural language processing model.

[0316] A "response" is the reply that the generative AI model creates in response to a message from a scammer.

[0317] "Sending" is the act of sending the generated response to the fraudster over the network.

[0318] "Storage" refers to the act of recording the content of interactions with a fraudster in a storage device such as a database.

[0319] "Extraction" is the process of extracting specific information from the content of stored interactions.

[0320] "Safe information" is information that has been determined to pose no risk to users as a result of interactions with fraudsters.

[0321] "User" means an individual or organization using this system who is to be protected from fraudulent activity by fraudsters.

[0322] The invention is a system that receives messages from fraudsters, analyzes them, generates appropriate responses, and sends them to the fraudsters. It also aims to store the results of interactions with fraudsters, extract secure information, and provide it to users.

[0323] Hardware and software used

[0324] Hardware: Servers, terminals

[0325] Software: Generative AI models (e.g., GPT-4), database management systems (e.g., MySQL®)

[0326] Specific operation of the system

[0327] Receiving messages from scammers

[0328] The server receives messages sent by fraudsters over the network. For example, a fraudster may send a message saying, "Please tell us your bank account information."

[0329] Message Parsing

[0330] The server inputs the received message into the generative AI model and analyzes its content. Specifically, the server inputs a prompt to the generative AI model: "Analyze the message from the scammer and determine whether it contains inappropriate requests or misleading information." The generative AI model then determines whether the message is inappropriate.

[0331] Generating a response

[0332] The server uses the generative AI model to generate an appropriate response to the fraudster. For example, the server inputs a prompt to the generative AI model saying, "Please generate and send an appropriate response to the fraudster." The generative AI model generates a response saying, "The request is inappropriate. We cannot provide personal information."

[0333] Sending a Response

[0334] The server sends the generated response to the fraudster. The server sends the response generated by the generative AI model to the fraudster over the network.

[0335] Saving the results of the interaction

[0336] The server stores the results of the interaction with the fraudster in a database. Specifically, the server stores the content of the interaction (messages from the fraudster, generated responses) in a database management system (e.g., MySQL).

[0337] Secure Information Extraction

[0338] The server analyzes the results of the interaction and extracts only the information that is deemed safe. The server inputs the content of the interaction stored in the database into the generative AI model again, and analyzes it using the prompt, "Analyze the results of the interaction with the scammer and extract only the information that is deemed safe."

[0339] Providing information to users

[0340] The server sends the information that has been determined to be safe to the user's terminal. The terminal displays the received information to the user. Specifically, the server sends the information that has been determined to be safe to the user's terminal via the network, and the terminal displays the received information on its screen and notifies the user.

[0341] Prompt Sentence Examples

[0342] "Analyze messages from scammers to determine if they contain inappropriate requests or misleading information."

[0343] "Generate and send an appropriate response to the scammer."

[0344] "Analyze the results of your interactions with the scammer and extract only the information that is deemed safe."

[0345] This system automates communication with fraudsters and provides only safe information to users, thereby preventing fraud damage. The flow of the identification process in the third embodiment will be described with reference to FIG.

[0346] Step 1:

[0347] Receiving messages from scammers

[0348] The server receives messages sent by fraudsters over the network.

[0349] Input: A message from the scammer (e.g., "Please tell me your bank details")

[0350] Output: Received message data

[0351] Specific operation: The server receives a message through the network interface and stores the message data in its internal memory.

[0352] Step 2:

[0353] Message Parsing

[0354] The server inputs the received message into a generative AI model and analyzes the content.

[0355] Input: Received message data

[0356] Output: Message analysis results (e.g., whether or not there are any inappropriate requests)

[0357] How it works: The server inputs message data into the generative AI model along with a prompt: "Analyze the message from the scammer and determine whether it contains inappropriate requests or misleading information." The generative AI model analyzes the content of the message and returns the results to the server.

[0358] Step 3:

[0359] Generating a response

[0360] The server uses a generative AI model to generate appropriate responses to the imposter.

[0361] Input: Message parsing result

[0362] Output: The generated response (e.g., "That request is inappropriate. We cannot give out personal information.")

[0363] Specific operation: The server inputs the prompt "Generate and send an appropriate response to the fraudster" and the analysis results to the generative AI model. The generative AI model generates an appropriate response and returns it to the server.

[0364] Step 4:

[0365] Sending a Response

[0366] The server generates a response and sends it to the impostor.

[0367] Input: The generated response

[0368] Output: Response transmission result (e.g. confirmation that the response was sent successfully)

[0369] Specific operation: The server sends the response generated by the generative AI model to the fraudster via the network, verifies the transmission result, and records it in a log.

[0370] Step 5:

[0371] Saving the results of the interaction

[0372] The server stores the results of the interaction with the fraudster in a database.

[0373] Input: Message from scammer, generated response, and submission result

[0374] Output: Save result (e.g., confirmation that the data was saved successfully to the database)

[0375] Specific operation: The server saves the contents of the exchange (message from the scammer, generated response, and transmission result) in a database management system (e.g., MySQL). It checks the saved results and records them in a log.

[0376] Step 6:

[0377] Secure Information Extraction

[0378] The server analyzes the results of the exchange and extracts only the information that is deemed safe.

[0379] Input: The conversation details stored in the database

[0380] Output: Secure information (e.g., information that can be provided to the user)

[0381] Specific operation: The server inputs the content of the interaction stored in the database into the generative AI model again, and analyzes it using the prompt, "Analyze the results of the interaction with the scammer and extract only the information that is determined to be safe." The generative AI model extracts the safe information and returns the results to the server.

[0382] Step 7:

[0383] Providing information to users

[0384] The server sends the information that has been determined to be safe to the user's terminal, which then displays the received information to the user.

[0385] Input: Secure Information

[0386] Output: Information displayed to the user

[0387] Specific operation: The server sends the information judged to be safe to the user's device via the network. The device displays the received information on its screen and notifies the user.

[0388] (Application example 3)

[0389] Next, a description will be given of Application Example 3 of Form Example 3. 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."

[0390] In recent years, there has been an increase in sophisticated fraud methods carried out via telephone and email, increasing the risk that users will be deceived by fraudsters' verbal tactics and documents. Elderly people and those with low digital literacy are particularly susceptible to fraud. In this situation, there is a need for a system that can automatically detect interactions with fraudsters and respond appropriately. It is also necessary to automatically generate appropriate responses to potentially fraudulent messages and provide users with only safe information.

[0391] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 3 is realized by the following means. In this invention, the server includes means for detecting anything unusual in the speech techniques or documents of a clever scammer who accesses the server by phone, email, etc., means for suggesting response content to uncover the scammer's content, means for interacting with and responding to the scammer, means for connecting the user to only those messages that are determined to be safe, means for automatically generating appropriate responses to messages that may be fraudulent, means for reporting the results of the interaction with the scammer to the user, and means for determining the possibility of fraud using a generative AI model and generating prompt text. This makes it possible to automatically detect interactions with scammers and respond appropriately.

[0392] The "means for detecting unusual content" is a function that automatically identifies unnatural expressions or requests that may be fraudulent in messages or call content received by telephone, email, etc.

[0393] The "means for proposing response content to detect the content of a fraudster" is a function for generating appropriate measures and responses to the information and requests provided by a fraudster.

[0394] "Means for interacting and responding to scammers" refers to a feature that automatically responds to messages and calls from scammers to reduce the risk of fraud.

[0395] "Means of connecting users with only those items that are deemed safe" is a function that notifies users of only information and requests that are deemed to pose no risk of fraud.

[0396] The "means for automatically generating an appropriate response to a potentially fraudulent message" is a function that automatically generates an appropriate response to a message that is suspected to be fraudulent.

[0397] "Means for reporting the results of interactions with scammers to users" is a function for reporting the content and results of interactions with scammers to users.

[0398] "Means for determining the likelihood of fraud and generating prompt sentences using a generative AI model" refers to a function that uses a generative AI model to assess the likelihood of fraud and create prompt sentences to generate appropriate responses.

[0399] The system for implementing this invention involves three parties working together: a server, a terminal, and a user. The server plays a central role in detecting interactions with fraudsters and generating appropriate responses. The terminal is a device through which the user receives messages and calls from fraudsters and communicates with the server. The user receives information provided by the system and takes action as necessary.

[0400] The server first uses a method to detect any unusual content in the speech techniques and documents of sophisticated fraudsters who access the server by phone, email, etc. This method uses natural language processing technology to analyze the contents of messages and calls and identify unnatural expressions and requests that may be fraudulent.

[0401] Next, a method is used to suggest responses to uncover fraudsters' content. This method uses a generative AI model to generate appropriate countermeasures and responses to the information and requests provided by fraudsters. The generative AI model has learned from past fraud cases and patterns, and is able to generate highly accurate responses.

[0402] Additionally, the system uses a means of communicating and responding to fraudsters, which automatically responds to messages and calls from fraudsters to mitigate the risk of fraud, for example by automatically generating appropriate responses to potentially fraudulent messages and responding to fraudsters.

[0403] In addition, a method is used to connect only those that are judged to be safe to the user. This method notifies the user only of information and requests that are judged to have no risk of fraud. This allows the user to receive only safe information.

[0404] The method also includes a means for reporting the results of interactions with fraudsters to the user. This means reports the content and results of interactions with fraudsters to the user, allowing the user to understand the situation.

[0405] For example, if a server receives a message from a fraudster saying, "Please tell us your bank account information," the server analyzes the message and determines that it is likely a scam. Using a generative AI model, the server automatically generates a response to the fraudster saying, "This request is inappropriate. Please ignore it." The user is then notified of the outcome of this interaction.

[0406] Examples of prompts include:

[0407] Message: "Please tell us your bank details."

[0408] Response: "This request is inappropriate. Please ignore it."

[0409] In this way, the server, terminal, and user can work together to reduce the risk of fraud and provide only safe information to the user.

[0410] The flow of the specific processing in Application Example 3 will be described with reference to FIG.

[0411] Step 1:

[0412] The server analyzes messages and call content received from the device. The input is the messages and call content received by the user. The server uses natural language processing technology to convert the messages and call content into text data and identify unnatural expressions and requests that may be fraudulent. The output is the analysis results.

[0413] Step 2:

[0414] The server determines whether there is a possibility of fraud based on the analysis results. The input is the analysis results obtained in step 1. The server detects keywords and patterns that may indicate fraud and assesses the risk of fraud. The output is a determination of whether there is a possibility of fraud.

[0415] Step 3:

[0416] If the server determines that there is a possibility of fraud, it uses a generative AI model to generate an appropriate response. The inputs are the fraud possibility determination result obtained in step 2, and the original message and call content. The server inputs a prompt sentence into the generative AI model to generate an appropriate response. The output is the generated response.

[0417] Step 4:

[0418] The server automatically sends the generated response to the impostor. The input is the response generated in step 3. The server sends an appropriate response to the impostor, mitigating the risk of fraud. The output is the response sent to the impostor.

[0419] Step 5:

[0420] The server notifies the user only of information or requests that are determined to be free of fraud risks. The input is the fraud possibility determination result obtained in step 2. The server notifies the user of information that is determined to be safe, allowing the user to receive only safe information. The output is the safe information that has been notified to the user.

[0421] Step 6:

[0422] The server reports the results of the interaction with the impostor to the user. The input is the content of the interaction with the impostor that took place in step 4. The server summarizes the results of the interaction and reports them to the user. The output is the results of the interaction reported to the user.

[0423] In this way, the server, terminal, and user can work together to reduce the risk of fraud and provide only safe information to the user.

[0424] 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.

[0425] "Example 1"

[0426] In one embodiment of the present invention, a system is provided in which a means for suggesting response content to detect fraudsters' messages is provided by generative AI, and further includes an emotion engine that recognizes the user's emotions. This system suggests response content to detect fraudsters' messages in response to information received by the user via telephone or email, and communicates and responds to fraudsters. Furthermore, the emotion engine analyzes emotions from the user's tone of voice and the text of emails, and provides the results to the system's means for communicating and responding to fraudsters.

[0427] "Example 2"

[0428] For example, if a user receives a call from a scammer, the system will detect something unusual about the scammer's speech and use generative AI to suggest a response that will expose the scammer's message. At the same time, the emotion engine will analyze the user's emotions from their tone of voice and provide the results to the system. The system will use this information to communicate and respond to the scammer.

[0429] "Example 3"

[0430] Additionally, if the emotion engine determines that the user's emotion is anxiety or fear, the system's means of interacting with and responding to the scammer will interrupt the interaction with the scammer and provide the user with safety information. For example, if the emotion engine determines that the user is feeling anxiety or fear in response to a call from a scammer, the system will automatically disconnect the call and notify the user that it was a call from a scammer and how to respond.

[0431] The processing flow of each embodiment will be described below.

[0432] "Example 1"

[0433] Step 1: The user receives information by phone or email.

[0434] Step 2: The system uses generative AI to suggest responses to identify the fraudster.

[0435] Step 3: The emotion engine analyzes the user's tone of voice and the content of the email to determine their emotions.

[0436] Step 4: The results of the emotion engine analysis are used to inform the system's interactions and responses to fraudsters.

[0437] "Example 2"

[0438] Step 1: The user receives a call from a scammer.

[0439] Step 2: The system detects anything that seems odd about the scammer's speech and uses generative AI to suggest a response that will uncover the scammer's content.

[0440] Step 3: The emotion engine analyzes the emotion from the user's tone of voice and provides the results to the system.

[0441] Step 4: The system interacts with and responds to the scammer based on the results of the emotion engine analysis.

[0442] "Example 3"

[0443] Step 1: The emotion engine determines that the user's emotion is anxiety or fear.

[0444] Step 2: The system's means of interacting with and dealing with the scammers stops interacting with them.

[0445] Step 3: The system provides the user with secure information, such as informing them that the call was from a scammer and how to deal with it.

[0446] Example 1

[0447] Next, a description will be given of Example 1 of Form 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."

[0448] In modern society, sophisticated frauds using telephone and email are on the rise, making it difficult for users to detect these scams. Furthermore, users are unable to respond appropriately when interacting with fraudsters, which can lead to increased damage. Furthermore, while it is necessary to respond in a way that takes into account the user's feelings, current systems do not adequately address this issue. Effective methods to resolve these issues are needed.

[0449] 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.

[0450] In this invention, the server includes means for detecting inconsistent content in the speech techniques and documents of sophisticated fraudsters who access the server by telephone, email, etc., means for proposing response content to detect the fraudster's content, means for communicating and responding to the fraudster, means for connecting only those determined to be safe to the user, means for analyzing the user's emotions, and means for adjusting the response content based on the analyzed emotions. This makes it possible to quickly detect fraudulent content and propose appropriate responses to the user. Furthermore, by providing response content that takes the user's emotions into consideration, an environment in which the user can respond with peace of mind can be provided.

[0451] "Telephone and email" refers to the means of sending and receiving information using voice communication and text messages.

[0452] A "sophisticated fraudster" is someone who commits fraud using sophisticated verbal and written techniques to deceive users.

[0453] "Means for detecting unusual content" refers to technology that automatically identifies fraudulent elements or unnatural information.

[0454] The "means for suggesting response content to detect fraudsters" is a technology that generates an appropriate response to detect fraudulent activity.

[0455] "Methods of interacting and responding to fraudsters" refers to techniques for interacting with fraudsters and taking appropriate action.

[0456] "Means of connecting users with only those things that are determined to be safe" is a technology that provides users with only information and communications that are determined to pose no risk of fraud.

[0457] "Means for analyzing user emotions" refers to technology that identifies emotions from the tone of a user's voice and the content of their text.

[0458] "Means for adjusting response content based on analyzed emotions" refers to a technology that optimizes response content according to the user's emotional state.

[0459] "Generative AI" is an AI technology that uses natural language processing and machine learning to generate text and speech.

[0460] The present invention provides a system that helps users respond appropriately to sophisticated fraudulent activities using telephone and email. The system includes multiple means for detecting fraudulent content and suggesting appropriate responses.

[0461] Hardware and software used

[0462] 1. Server

[0463] Speech Recognition Software: Google® Cloud Speech-to-Text API is used to convert received voice data into text.

[0464] Natural language processing software: spaCy is used to analyze text data and extract grammar and meaning.

[0465] Generative AI model: OpenAI GPT-4 is used to detect fraudulent elements and generate appropriate response content.

[0466] Sentiment Analysis Engine: Uses IBM Watson (registered trademark) Tone Analyzer to analyze the user's emotions.

[0467] 2. Terminal

[0468] Voice recording function: Records the voice of the call and sends it to the server.

[0469] Text sending function: Sends the contents of received emails to the server.

[0470] Display function: Presents the generated response content to the user.

[0471] 3. Users

[0472] Receiving phone calls or emails: You may receive contact from scammers.

[0473] Review responses: Review the responses provided and communicate with the scammer.

[0474] Specific operation of the system

[0475] 1. Receiving Data

[0476] The user receives a phone call or email.

[0477] The terminal transmits the received voice data and text data to the server.

[0478] 2. Data Analysis

[0479] The server parses the received data.

[0480] For audio data, the server converts the audio to text using the Google Cloud Speech-to-Text API.

[0481] For text data, the server uses spaCy to parse the document and extract grammar and semantics.

[0482] 3. Fraud detection

[0483] The text data analyzed by the server is input into a generative AI model (OpenAI GPT-4) to determine whether it contains any fraudulent elements.

[0484] The server detects certain keywords and phrases (e.g., "bank account," "fraud") and assesses the likelihood of fraud.

[0485] 4. Emotion Analysis

[0486] The device inputs the user's tone of voice and the text of the email into an emotion engine (IBM Watson Tone Analyzer) to analyze the user's emotions.

[0487] The device sends the results of the emotion analysis to the server.

[0488] 5. Response suggestions

[0489] The server uses a generative AI model to generate appropriate response content to detect the fraudster's content.

[0490] The server takes into account the results of the emotion analysis and adjusts the response content to make the user feel at ease.

[0491] 6. Providing a Response

[0492] The terminal presents the generated response to the user.

[0493] The user uses the provided responses to interact with the fraudster.

[0494] Specific examples

[0495] For example, if a user receives an email saying, "Your bank account has been fraudulently used. Please contact us immediately," the system operates as follows:

[0496] 1. Receiving Data

[0497] The user receives the email and the terminal sends the contents to the server.

[0498] 2. Data Analysis

[0499] The server analyzes the text of the email and detects any fraudulent elements.

[0500] 3. Fraud detection

[0501] The server detects the phrase "fraudulent use of bank account" and determines that it is likely a scam.

[0502] 4. Emotion Analysis

[0503] The device analyzes the user's emotions and transmits to the server that the user is feeling anxious.

[0504] 5. Response suggestions

[0505] The server uses the generative AI model to generate a response that reads, "This email may be fraudulent. We recommend that you contact the bank directly through their official website."

[0506] 6. Providing a Response

[0507] The terminal presents the generated response content to the user, and the user acts based on the content.

[0508] Prompt Sentence Examples

[0509] An example of a prompt sentence to be input to the generative AI model is as follows.

[0510] I received an email saying "Your bank account has been compromised. Please contact us immediately." Please help me determine if this email is a scam and suggest an appropriate response.

[0511] In this way, the system helps users respond appropriately to the sophisticated verbal and written techniques of fraudsters.

[0512] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0513] Step 1:

[0514] Receiving data

[0515] The user receives a phone call or email.

[0516] The voice data and text data received by the terminal are sent to the server.

[0517] Input: Telephone voice data, email text data.

[0518] Output: Audio and text data sent to the server.

[0519] Specific operation: The device records the audio of the phone call and sends the contents of the email directly to the server.

[0520] Step 2:

[0521] Data analysis

[0522] The server parses the received data.

[0523] For audio data, the server converts the audio to text using the Google Cloud Speech-to-Text API.

[0524] For text data, the server uses spaCy to parse the document and extract grammar and semantics.

[0525] Input: Audio and text data sent to the server.

[0526] Output: Parsed text data.

[0527] What it does: The server converts the voice data into text and passes the text to a natural language processing engine for analysis.

[0528] Step 3:

[0529] Fraud detection

[0530] The text data analyzed by the server is input into a generative AI model (OpenAI GPT-4) to determine whether it contains any fraudulent elements.

[0531] The server detects certain keywords and phrases (e.g., "bank account," "fraud") and assesses the likelihood of fraud.

[0532] Input: Parsed text data.

[0533] Output: A fraud likelihood score.

[0534] How it works: The server inputs text data into a generative AI model and receives a score indicating whether it is likely to be fraudulent.

[0535] Step 4:

[0536] Emotion Analysis

[0537] The device inputs the user's tone of voice and the text of the email into an emotion engine (IBM Watson Tone Analyzer) to analyze the user's emotions.

[0538] The device sends the results of the emotion analysis to the server.

[0539] Input: User's tone of voice, email text.

[0540] Output: Parsed emotion data.

[0541] Specific operation: The device passes voice and text data to the emotion engine and sends the analysis results to the server.

[0542] Step 5:

[0543] Response suggestions

[0544] The server uses a generative AI model to generate appropriate response content to detect the fraudster's content.

[0545] The server takes into account the results of the emotion analysis and adjusts the response content to make the user feel at ease.

[0546] Input: Fraud likelihood score, parsed sentiment data.

[0547] Output: The generated response.

[0548] Specific operation: The server inputs a prompt sentence into the generative AI model and fine-tunes the generated response content based on the results of emotion analysis.

[0549] Step 6:

[0550] Providing a response

[0551] The terminal presents the generated response to the user.

[0552] The user uses the provided responses to interact with the fraudster.

[0553] Input: The generated response content.

[0554] Output: The response presented to the user.

[0555] Specific operation: The device displays the response on the screen, and the user confirms the content.

[0556] (Application example 1)

[0557] Next, a description will be given of Application Example 1 of Embodiment 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."

[0558] There is an increasing risk that users will easily be deceived by the persuasive tactics and written messages of scammers who contact them cleverly via phone or email. Particularly problematic are the clever tactics used by scammers that lead users to respond to inappropriate requests or be misled by misleading information. Furthermore, users often feel anxiety and stress about scams, and appropriate responses to these issues are required. To solve these issues, a system is needed that automates fraud detection and response, and analyzes users' emotions to provide appropriate support.

[0559] 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.

[0560] In this invention, the server includes means for detecting anything unusual in the speech techniques or documents of clever scammers who access the server by phone or email, means for suggesting responses to uncover the scammer's content, means for communicating and responding to the scammer, means for analyzing the user's emotions, and means for connecting the user only with connections that are determined to be safe. This reduces the risk of fraud and allows users to communicate with peace of mind.

[0561] "Means for detecting unusual content" refers to technology that detects unnatural or inappropriate content in the speech techniques or documents of sophisticated fraudsters who access you via phone or email.

[0562] The "means for suggesting response content to detect fraudsters' content" is a technology for generating and suggesting appropriate responses to inappropriate requests or misleading information that fraudsters make to users.

[0563] "Means for communicating and responding to fraudsters" refers to technology that automatically responds to contact from fraudsters and takes appropriate action.

[0564] "Means for analyzing user emotions" refers to technology that analyzes emotions from the tone of a user's voice or the text of an email, and provides the results to other functions of the system.

[0565] "Means of connecting users only with communications that are determined to be safe" is a technology that provides users with only communications that are determined to be void of fraud risks.

[0566] "Generative AI" is an artificial intelligence technology that automatically generates responses to detect fraudsters' messages.

[0567] An "emotion engine" is a technology that analyzes the user's emotions and provides the results to other functions of the system.

[0568] As an embodiment of this invention, we will explain how to install a fraud detection assistant system on a smartphone. This system has the functions of detecting anything unusual in the speech techniques and documents of sophisticated fraudsters who access the smartphone via phone or email, suggesting responses to uncover the fraudster's content, and analyzing the user's emotions.

[0569] Hardware and software used

[0570] Hardware: Smartphone

[0571] Software: Python (registered trademark), regular expression library, sentiment analysis library, generative AI library

[0572] System configuration

[0573] 1. How to detect inappropriate content:

[0574] The server uses a regular expression library to analyze the content of calls and emails to detect whether they contain fraudulent patterns, thereby identifying messages that pose a risk of fraud.

[0575] 2. Suggested response methods to detect scammers:

[0576] The server uses a generative AI library to generate appropriate responses to the scammer's content, providing guidance to users on how to respond appropriately to the scammer.

[0577] 3. How to communicate and respond to scammers:

[0578] The server automatically responds to contact from fraudsters and takes appropriate action, allowing users to safely communicate with fraudsters.

[0579] 4. Means of analyzing user sentiment:

[0580] The server uses an emotion analysis library to analyze the emotion of the user's tone of voice and the content of the email, allowing it to provide additional support if the user is feeling stressed or anxious.

[0581] 5. Methods to connect users only with devices that are deemed safe:

[0582] The server will only provide users with communications that are deemed to be fraud-free, allowing users to communicate with peace of mind.

[0583] Specific examples

[0584] For example, if a user receives a message saying, "Please provide your bank account information. Urgent action is required," the system will operate as follows:

[0585] 1. Detecting unusual content:

[0586] The server uses a regular expression library to parse the message and detect that the phrase "Please tell me your bank account details" matches a fraudulent pattern.

[0587] 2. Response suggestions:

[0588] The server uses a generative AI library to generate an appropriate response: "Sorry, we cannot provide any personal information."

[0589] 3. Emotion analysis:

[0590] The server uses an emotion analysis library to analyze whether the user is feeling stressed by this message.

[0591] 4. Providing secure communications:

[0592] The server determines that this message poses a risk of fraud and displays a warning to the user.

[0593] Prompt Sentence Examples

[0594] message = "Please provide your bank details. This is urgent."

[0595] result = assistant.handle_message(message)

[0596] print(result)

[0597] In this way, the fraud detection assistant system helps users reduce the risk of being scammed and enables them to communicate with peace of mind.

[0598] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0599] Step 1:

[0600] The server receives phone and email messages. As input, it provides the message received by the user. The server uses a regular expression library to parse this message. Specifically, it scans the text in the message for phrases that match fraud patterns. As output, it generates a flag indicating whether the message is suspected of fraud.

[0601] Step 2:

[0602] If the server determines that fraud is suspected, it uses a generative AI library to generate an appropriate response. The server receives a suspected fraudulent message as input. Based on this message, the server generates an appropriate response to the fraudster. The generated response is the output.

[0603] Step 3:

[0604] The server uses a sentiment analysis library to analyze the user's sentiment. As input, it receives a message from the user. The server analyzes the text and audio data of the message to assess the level of stress and anxiety felt by the user. As output, it obtains the results of the sentiment analysis.

[0605] Step 4:

[0606] The server displays a warning to the user based on the suspected fraudulent message and the results of sentiment analysis. As input, the server receives the suspected fraudulent message and the results of sentiment analysis. Based on this information, the server notifies the user of the risk of fraud and prompts them to take appropriate action. As output, a warning message is generated that is displayed to the user.

[0607] Step 5:

[0608] The server provides only messages that are determined to be safe to the user. As input, messages that are determined to be free of fraud are provided. The server forwards these messages to the user, allowing them to communicate with confidence. As output, safe messages are provided to the user.

[0609] In this way, the fraud detection assistant system helps users reduce the risk of being scammed and enables them to communicate with peace of mind.

[0610] Example 2

[0611] Next, a description will be given of Example 2 of Form 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."

[0612] Conventional systems have difficulty responding appropriately to the sophisticated speech techniques and documents of fraudsters, putting users at high risk of being scammed. Furthermore, they are unable to take into consideration the user's feelings when interacting with fraudsters, often resulting in stress and anxiety. This has led to a need for effective measures to ensure user safety.

[0613] 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.

[0614] In this invention, the server includes a means for detecting any inconsistencies in the speech techniques and documents of sophisticated scammers who access the server by phone, email, etc., a means for proposing response content to uncover the scammer's content, a means for communicating and responding to the scammer, a means for connecting only those determined to be safe to the user, a means for analyzing emotions from the user's tone of voice, and a means for proposing an optimal response based on the generated response content and the emotion analysis results. This makes it possible to respond appropriately to the scammer's clever speech techniques and documents and ensure the user's safety. Furthermore, by taking the user's emotions into consideration when responding, stress and anxiety can be reduced.

[0615] "Telephone, email, etc." refers to means of communication such as voice calls and email.

[0616] A "fraudster" is someone who attempts to fraudulently obtain money or information by deceiving others.

[0617] "Speaking" refers to the skill and techniques of using words skillfully.

[0618] "Document" refers to written information or messages.

[0619] "Unusual content" refers to unnatural and suspicious content that differs from normal communication.

[0620] "Means for sensing" refers to a method or device for identifying and detecting a particular condition or pattern.

[0621] "Response content" refers to the content of a reply to the other person's words or actions.

[0622] "Suggestion means" refers to a method or device for presenting a particular action or option.

[0623] "Means of communication and response" refers to the methods and devices used to communicate and respond to others.

[0624] "Things that have been determined to be safe" refers to information or communications that are deemed to be low risk and safe for users.

[0625] "User" refers to any individual or entity that uses the System.

[0626] "Tone of voice" refers to the pitch, intensity, and emotional expression of the voice.

[0627] "Means for analyzing emotions" refers to methods and devices for identifying and analyzing emotions from voice, facial expressions, etc.

[0628] "Generated response content" refers to a response content automatically generated by the system.

[0629] "Emotion analysis results" refers to information obtained as a result of analyzing emotions.

[0630] The "means for suggesting an optimal response" refers to a method or device for suggesting the most appropriate response depending on the situation.

[0631] "Generative AI" refers to AI that has the ability to generate new information and responses based on data.

[0632] "Voice and email services" refers to communications services that provide voice calls and email.

[0633] "Option Menu" refers to selectable functions and services provided in addition to the basic functions.

[0634] The present invention provides a system for ensuring the safety of users by appropriately dealing with the clever speech techniques and documents of fraudsters. A specific embodiment of this system will be described below.

[0635] Hardware and software used

[0636] Hardware: Servers, devices (smartphones and PCs)

[0637] Software: Generative AI models (e.g., GPT-4), emotion engines

[0638] System Overview

[0639] The system analyzes the scammer's speaking style and the content of the document, generates appropriate responses, and provides them to the user. Specifically, it suggests ignoring or denying responses to inappropriate requests or misleading information from the scammer.

[0640] Data processing and calculation flow

[0641] 1. The user receives a call from a scammer.

[0642] 2. The device records the call and sends it to the server in real time.

[0643] 3. The server receives the recording data and inputs it into the generative AI model.

[0644] 4. The generative AI model detects anything that seems odd about the scammer's speech and generates an appropriate response.

[0645] 5. The emotion engine analyzes the emotion from the tone of the user's voice and provides the results to the server.

[0646] 6. The server proposes the optimal response to the user based on the generated response content and the results of sentiment analysis.

[0647] Specific examples

[0648] Scenario: A user is asked by a scammer to "provide your bank account details."

[0649] An example prompt for a generative AI model: "If a scammer is asking for your bank account information, suggest how you should respond."

[0650] Implementation details

[0651] When a user receives a call from a scammer, the device records the conversation and sends it to a server in real time.

[0652] The server receives the recorded data and inputs it into a generative AI model (e.g., GPT-4), which detects anything inconsistent with the scammer's speech and generates a response such as, "The request is inappropriate. Please do not provide any personal information."

[0653] At the same time, the emotion engine analyzes the user's emotions from the tone of their voice and provides the server with results such as "the user is feeling anxious."

[0654] The server then proposes the best response to the user based on the generated response and the results of sentiment analysis, providing specific advice such as "Please ignore the request. Do not provide personal information."

[0655] In this way, the system can detect the fraudster's behavior and suggest appropriate responses to the user, ensuring user safety and preventing fraud before it occurs. It also takes into account the user's emotions, reducing stress and anxiety.

[0656] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0657] Step 1:

[0658] A user receives a call from a scammer.

[0659] What happens: The user receives a call on their smartphone in which a scammer asks for their bank account details.

[0660] Enter: A call from a scammer.

[0661] Output: Audio data of the call.

[0662] Step 2:

[0663] The device records the call and sends it to the server in real time.

[0664] Specific operation: The smartphone's recording function is activated, and the call is recorded while streaming the data to the server.

[0665] Input: Audio data from the call.

[0666] Output: The audio data sent to the server.

[0667] Step 3:

[0668] The server inputs the recorded data into a generative AI model.

[0669] How it works: The server analyzes the recording, converts the scammer's speech into text data, and feeds it into a generative AI model (e.g., GPT-4).

[0670] Input: The audio data sent to the server.

[0671] Output: The text data fed into the generative AI model.

[0672] Step 4:

[0673] The generative AI model generates appropriate response content.

[0674] Specific behavior: The generative AI model generates a response such as, "That request is inappropriate. Please do not provide personal information."

[0675] Input: Text data fed into the generative AI model.

[0676] Output: The generated response.

[0677] Step 5:

[0678] The emotion engine analyzes emotions from the user's tone of voice.

[0679] Specific operation: The emotion engine analyzes the tone of the user's voice and sends the emotion analysis result, such as "the user is feeling anxious," to the server.

[0680] Input: Audio data from the call.

[0681] Output: Emotion analysis results.

[0682] Step 6:

[0683] The server proposes the optimal response to the user based on the generated response content and the results of sentiment analysis.

[0684] Specific action: The server notifies the user by sending specific advice to the user's device, such as "Please ignore the request. Do not provide personal information."

[0685] Input: Generated response content, sentiment analysis results.

[0686] Output: The best response suggested to the user.

[0687] (Application example 2)

[0688] Next, a description will be given of Application Example 2 of Form 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."

[0689] In recent years, sophisticated fraudulent activities using telephone and email have been on the rise, increasing the risk that users will be deceived by the scammers' verbal tactics and written documents. Furthermore, delayed responses to fraudulent activities can increase users' mental stress and anxiety. Conventional systems struggle to detect fraudulent activities in real time and generate appropriate responses, and are also inadequate in responding while taking into account the user's emotional state. Therefore, there is a need for a system that can reduce the mental burden on users while ensuring their safety.

[0690] 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.

[0691] In this invention, the server includes a means for detecting anything unusual in the speech techniques or documents of sophisticated fraudsters who access the server by phone or email, a means for suggesting responses to uncover the fraudster's content, a means for communicating and responding to the fraudster, a means for analyzing the user's emotional state and providing advice to reduce stress and anxiety, and a means for connecting the user only to those connections that are determined to be safe. This makes it possible to detect fraudulent activity in real time and generate appropriate responses. It also makes it possible to respond in consideration of the user's emotional state, thereby reducing mental burden.

[0692] "Telephone, email, etc." refers to means of communication between users and the outside world, including voice communication and email.

[0693] A "sophisticated fraudster" is someone who commits fraud using sophisticated verbal and written techniques to deceive users.

[0694] "Perlocution and writing" refers to the words and written content that scammers use to deceive users.

[0695] "Methods for detecting unusual content" are techniques for identifying unnatural or suspicious elements in a fraudster's speech or documents.

[0696] The "means for suggesting response content to detect the content of a fraudster" is a technology for generating appropriate responses to the speech or writing of a fraudster and providing them to the user.

[0697] "Methods of interacting with and responding to fraudsters" are techniques for managing communication with fraudsters and responding appropriately.

[0698] "Means for analyzing the user's emotional state and providing advice to reduce stress and anxiety" refers to technology that analyzes the user's emotions and provides advice to reduce mental burden.

[0699] "Means of connecting users only with communications that are determined to be safe" is a technology that connects users only with communications that are determined to pose no risk of fraud.

[0700] "Generative AI" is a technology that uses artificial intelligence to generate text and speech.

[0701] "Emotion analysis" is a technology that identifies emotions from a user's tone of voice and text.

[0702] The system for implementing this invention detects any inconsistencies in the speech techniques or documents of sophisticated scammers who access the user via phone or email, and suggests responses to uncover the scammer's content. Furthermore, the system interacts with and responds to the scammer, analyzes the user's emotional state, and provides advice to reduce stress and anxiety. It also has the function of connecting the user only to those connections that are determined to be safe.

[0703] Hardware and software used

[0704] Hardware: Smartphone (microphone, speaker)

[0705] software:

[0706] Python

[0707] Speech Recognition Library (Speech Recognition)

[0708] Transformers library (generative AI models)

[0709] TextBlob library (sentiment analysis)

[0710] System processing overview

[0711] 1. Speech recognition: The device (smartphone) uses a microphone to record voice and converts the voice into text using a speech recognition library.

[0712] 2. Fraud detection: The server checks whether the converted text contains certain keywords (e.g., bank account, password).

[0713] 3. Sentiment Analysis: The server analyzes the sentiment of the text using the TextBlob library. If the sentiment is negative, it determines that the likelihood of fraud is high.

[0714] 4. Response generation: The server uses a generative AI model (GPT-3 (registered trademark)) to generate an appropriate response.

[0715] 5. Displaying the results: The terminal displays the generated response to the user.

[0716] Specific examples

[0717] If a user receives a call from a scammer asking for their bank account information, the system will detect this and generate a response saying, "This may be a scam. Please ignore."

[0718] Prompt Sentence Examples

[0719] When receiving a call from a scammer, the following prompt is fed into the generative AI model:

[0720] "This may be a scam. Please ignore it."

[0721] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0722] Step 1:

[0723] The device (smartphone) uses a microphone to record the user's voice. The recorded voice data becomes the input. The device converts this voice data into text using a speech recognition library. The converted text becomes the output.

[0724] Step 2:

[0725] The server receives the text obtained in step 1 and performs fraud detection. Specifically, it checks whether the text contains specific keywords (e.g., bank account, password). If any of the keywords are detected, the text is determined to be potentially fraudulent. This determination result is output.

[0726] Step 3:

[0727] The server performs sentiment analysis on the text determined to be potentially fraudulent in step 2 using the TextBlob library. The input is the text from step 2, and the analysis results are a score of positive or negative sentiment. If the sentiment is negative, it is determined to be highly likely to be fraudulent.

[0728] Step 4:

[0729] The server generates an appropriate response using a generative AI model (GPT-3) based on the emotion analysis results from step 3. The input is the emotion analysis results from step 3 and text, and the generated response is the output. For example, a response such as "This may be a scam. Please ignore." is generated.

[0730] Step 5:

[0731] The terminal displays the response content generated in step 4 to the user. The input is the response content of step 4, which is presented to the user visually or audibly. The user can confirm this response content and take appropriate action.

[0732] Example 3

[0733] Next, a description will be given of a third embodiment of the third embodiment. 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."

[0734] Fraudulent acts by fraudsters using sophisticated speech or documents can cause serious damage to users. Conventional systems have had difficulty detecting fraudulent acts by fraudsters and responding appropriately. In addition, there was a lack of means to respond quickly when users felt anxious or scared. This made it difficult to ensure the safety of users.

[0735] The specific processing by the specific processing unit 290 of the data processing device 12 in the third embodiment is realized by the following means.

[0736] In this invention, the server includes means for receiving communications from fraudsters, means for analyzing the messages and determining the emotional state, means for generating a response to the fraudster, means for sending the response and reporting the results, means for providing only safe information to the user, and means for interrupting the interaction via the emotion engine. This allows fraudulent activity by fraudsters to be detected quickly and dealt with appropriately. Furthermore, if the user feels anxious or scared, a quick response can be made to ensure the user's safety.

[0737] "Means for receiving communications from scammers" refers to functions for receiving communications such as phone calls, emails, and messages from scammers.

[0738] The "means for analyzing messages and determining emotional state" is a function for analyzing received messages and determining the user's emotional state (anxiety, fear, etc.).

[0739] The "means for generating a response to an imposter" is a function for generating an appropriate response to an imposter.

[0740] The "means for sending a response and reporting the result" is a function for sending the generated response to the fraudster and reporting the result to the user.

[0741] "Means of providing only safe information to users" is a function that analyzes the content of interactions with scammers and provides users with only information that is deemed safe.

[0742] "Means to interrupt interaction using emotion engine" is a function that interrupts interaction with a scammer if the user's emotions are judged to be anxiety or fear.

[0743] A "generative AI model" is an artificial intelligence model used to generate responses to fraudsters.

[0744] This invention is a system for receiving communications from fraudsters and responding appropriately. The system is implemented using a server, terminals (PCs and smartphones), a generative AI model, an emotion engine, and a communication module.

[0745] Hardware and software used

[0746] Hardware: Servers, terminals (PCs, smartphones)

[0747] Software: Generative AI models (e.g., GPT-4), emotion engines, and communication modules

[0748] Specific operation of the system

[0749] 1. Receiving communications from scammers

[0750] The server receives communications (phone calls, emails, messages, etc.) from the fraudster.

[0751] The server logs the communications it receives.

[0752] Example: When a server receives a call from a scammer, it records the call and stores it in a database.

[0753] 2. Message analysis and emotional state determination

[0754] The server sends the data to the emotion engine for analysis of the received message.

[0755] The emotion engine analyzes the content of the message and determines the user's emotional state (anxiety, fear, etc.).

[0756] Example: The server sends a message from a scammer to the emotion engine, which determines that the user is feeling anxious.

[0757] 3. Generating responses to impostors

[0758] The server sends prompts to the generative AI model, which generates appropriate responses to the imposter.

[0759] The generative AI model generates a response based on the prompt sentence and sends it back to the server.

[0760] Example: The server sends the generative AI model a prompt saying, "Generate an appropriate response to a message from a scammer. Message content: 'Please tell us your bank account information.'" The generative AI model then generates a response saying, "We are unable to provide such information."

[0761] 4. Sending Responses and Reporting Results

[0762] The server sends the generated response to the impostor.

[0763] The server reports the results of the interaction to the user.

[0764] Example: The server generates a response and sends it to the fraudster, then notifies the user that it has responded appropriately to the fraudster's request.

[0765] 5. Providing safe information

[0766] The server analyzes the content of the exchange with the scammer and determines whether it is safe.

[0767] The server provides users with only information that is deemed safe.

[0768] Example: If the server analyzes a message from a scammer and determines that the information is safe, it notifies the user of that information.

[0769] 6. Emotional Engine Interrupts Interactions

[0770] If the emotion engine determines that the user's emotions are anxiety or fear, it will discontinue the interaction with the scammer.

[0771] The server automatically disconnects the call with the scammer and notifies the user that the call was from a scammer and how to deal with the situation.

[0772] Example: If the emotion engine determines that the user is feeling scared, the server will disconnect the call and notify the user that the call was from a scammer and that they should not answer calls from this number in the future.

[0773] Prompt Sentence Examples

[0774] "Generate an appropriate response to the scammer's message: 'Please provide your bank details.'"

[0775] "Please analyze the results of your interactions with scammers and only provide safe information to your users."

[0776] "Generate ways for users to respond if they feel anxious or scared."

[0777] This system provides advanced technology to automate interactions with fraudsters and ensure user safety. The flow of the identification process in the third embodiment will be described with reference to FIG.

[0778] Step 1:

[0779] Receiving communications from scammers

[0780] The server receives communications (phone calls, emails, messages, etc.) from the fraudster.

[0781] Input: Communication data from the scammer (voice, text, etc.)

[0782] Data processing: Log communication data and convert voice data to text as needed.

[0783] Output: Communication data in a parsable format

[0784] What it does: When the server receives a call from a scammer, it records the call, converts the audio data into text, and stores it in a database.

[0785] Step 2:

[0786] Message analysis and emotional state determination

[0787] The server sends the data to the emotion engine for analysis of the received message.

[0788] Input: Communication data in a parsable format

[0789] Data calculation: The emotion engine analyzes the content of the message and determines the user's emotional state (anxiety, fear, etc.).

[0790] Output: Emotional state judgment result

[0791] Specific operation: The server sends a message from the scammer to the emotion engine, which determines that the user is feeling anxious.

[0792] Step 3:

[0793] Generating responses to scammers

[0794] The server sends prompts to the generative AI model, which generates appropriate responses to the imposter.

[0795] Input: Emotional state judgment result, prompt sentence

[0796] Data computation: A generative AI model generates a response based on the prompt.

[0797] Output: The generated response

[0798] Specific operation: The server sends the generative AI model a prompt saying, "Generate an appropriate response to the message from the scammer. Message content: 'Please tell us your bank account information.'" The generative AI model then generates a response saying, "We are unable to provide such information."

[0799] Step 4:

[0800] Sending responses and reporting results

[0801] The server sends the generated response to the impostor.

[0802] Input: The generated response

[0803] Data manipulation: Send the response to the fraudster and log the results.

[0804] Output: Response sent to scammer, report to user

[0805] Specific behavior: The server sends the generated response to the fraudster, and then notifies the user that "an appropriate response has been provided to the fraudster's request."

[0806] Step 5:

[0807] Providing safe information

[0808] The server analyzes the content of the exchange with the scammer and determines whether it is safe.

[0809] Input: Your conversation with the scammer

[0810] Data calculation: Analyzes the content of the exchange and extracts information deemed safe.

[0811] Output: Information deemed safe

[0812] Specific operation: The server analyzes the message from the scammer and, if it determines that the information is safe, notifies the user of that information.

[0813] Step 6:

[0814] Emotion engine interrupts interaction

[0815] If the emotion engine determines that the user's emotions are anxiety or fear, it will discontinue the interaction with the scammer.

[0816] Input: User's emotional state

[0817] Data calculation: The emotion engine analyzes the user's emotional state and determines that they are feeling anxious or scared.

[0818] Output: Instruction to terminate the conversation, notification to the user

[0819] Specific behavior: If the emotion engine determines that the user is feeling scared, the server will disconnect the call and notify the user that the call was from a scammer and that they should not answer calls from this number in the future.

[0820] (Application example 3)

[0821] Next, a description will be given of Application Example 3 of Form Example 3. 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."

[0822] In modern society, sophisticated fraudulent activities via telephone and email are on the rise, increasing the risk that users will be deceived by fraudsters' verbal tactics and documents. In addition, users often feel anxious and scared when interacting with fraudsters, which places a heavy mental burden on them. In response to this, an effective system is needed to automatically detect fraudsters' inappropriate requests and misleading information and protect users.

[0823] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 3 is realized by the following means.

[0824] In this invention, the server includes means for detecting anything unusual in the speech techniques or documents of sophisticated scammers who access the server by phone, email, etc., means for suggesting responses to uncover the scammer's content, means for communicating and responding to the scammer, means for connecting the user only with information that is determined to be safe, means for analyzing the user's emotions and detecting anxiety or fear, means for interrupting communication with the scammer and providing safe information, means for evaluating the safety of information from the scammer, and means for generating appropriate responses to the scammer. This makes it possible to automatically detect attacks from scammers and protect users.

[0825] "Means for detecting unusual content" refers to technology that automatically detects unnatural points or inappropriate information contained in a fraudster's speech or documents.

[0826] "Means for suggesting response content to detect the intent of a fraudster" is a technology for detecting the intent of a fraudster and generating an appropriate response.

[0827] "Means of communicating and responding to scammers" refers to technology that communicates with scammers on behalf of the user and takes appropriate action on their behalf.

[0828] "Means of connecting users with only information that has been determined to be safe" is a technology that evaluates the safety of information from scammers and provides users with only information that has been determined to be safe.

[0829] "Means for analyzing the user's emotions and detecting anxiety or fear" refers to technology that analyzes the user's voice and facial expressions to determine whether the user is feeling anxiety or fear.

[0830] "Means to interrupt communication with scammers and provide safe information" refers to technology that automatically interrupts communication with scammers if the user feels anxious or scared, and provides the user with information that will reassure them.

[0831] The "means for assessing the safety of information from fraudsters" is a technique for assessing whether information provided by fraudsters is safe.

[0832] The "means for generating an appropriate response to an imposter" is a technique for automatically generating an appropriate response to an imposter.

[0833] The system for carrying out the present invention includes a series of means for automatically detecting calls and emails from fraudsters and protecting users. A specific embodiment of this system will be described below.

[0834] Hardware and software used

[0835] Hardware: Smartphones, smart glasses

[0836] Software: emotion engines, generative AI models (e.g., GPT-4), speech recognition systems (e.g., Google Speech-to-Text)

[0837] Data processing and calculation

[0838] 1. Speech Recognition:

[0839] The server converts calls and messages from scammers into text using a voice recognition system.

[0840] Usage example: Converting speech to text using the Google Speech-to-Text API.

[0841] 2. Sentiment analysis:

[0842] The server uses an emotion engine to analyze the user's voice and facial expressions to detect anxiety or fear.

[0843] Use case: Analyzing user sentiment in real time using a sentiment engine.

[0844] 3. Response generation using generative AI models:

[0845] The server uses a generative AI model (e.g., GPT-4) to generate appropriate responses to the imposter.

[0846] Usage example: Prompt GPT-4: "Generate a response that ignores requests from scammers."

[0847] 4. Safety assessment:

[0848] The server determines whether the information from the fraudster is safe and provides only information that is deemed safe to the user.

[0849] Use case: Using generative AI models to assess the safety of information from fraudsters.

[0850] 5. Notification and Action:

[0851] The server will interrupt the conversation with the scammer and notify the user that the call was from a scammer and how to deal with the situation.

[0852] Example use: Use the smartphone notification feature to display a warning message to the user.

[0853] Specific examples

[0854] Example prompts to input to the generative AI model

[0855] "Generate a response that ignores requests from scammers"

[0856] "Evaluate the safety of the information provided by the scammer"

[0857] Example scenario

[0858] 1. Speech Recognition:

[0859] When you receive a call from a scammer, your smartphone converts the voice to text.

[0860] For example: "Hello, your bank account is at risk."

[0861] 2. Sentiment analysis:

[0862] If the user is feeling anxious, the emotion engine will detect it.

[0863] Example: The emotion engine detects that the user's voice is trembling.

[0864] 3. Response generation using generative AI models:

[0865] The server uses the generative AI model to generate appropriate responses to the impostor.

[0866] For example: "That information is incorrect. Please check with your bank directly."

[0867] 4. Safety assessment:

[0868] The server evaluates the security of the information from the fraudster and determines that it is not secure.

[0869] For example, the information "Your bank account is at risk" is considered inappropriate.

[0870] 5. Notification and Action:

[0871] The server will interrupt the conversation with the scammer and send a warning message to the user.

[0872] For example: "This was a scammer calling. Please check with your bank directly."

[0873] In this way, the fraud prevention assistant can automatically detect attacks from fraudsters and protect users.

[0874] The flow of the specific processing in Application Example 3 will be described with reference to FIG.

[0875] Step 1:

[0876] The server receives calls and messages from scammers. It takes voice and text data as input, converts the voice data into text using a speech recognition system (e.g., Google Speech-to-Text), and obtains text data as output.

[0877] Step 2:

[0878] The server inputs the text data into an emotion engine to analyze the user's emotions. It receives the text data and the user's voice and facial expression data as input and detects anxiety or fear. It outputs the emotion analysis results.

[0879] Step 3:

[0880] The server inputs the text data into a generative AI model (e.g., GPT-4) to generate an appropriate response to the fraudster. Using the text data and a prompt (e.g., "Generate a response that ignores the fraudster's request") as input, the generative AI model generates a response. The response text is obtained as output.

[0881] Step 4:

[0882] The server evaluates the safety of the information from the fraudster before providing the generated response text to the user. Using the generated response text and the original text data from the fraudster as input, the server evaluates the safety using a generative AI model. The server obtains the safety evaluation result as output.

[0883] Step 5:

[0884] The server provides the response text to the user only if the information from the fraudster is deemed safe based on the safety evaluation result.The server uses the safety evaluation result and the response text as input and notifies the user of the information that is deemed safe.The server obtains the safe information that is provided to the user as output.

[0885] Step 6:

[0886] The server interrupts communication with the fraudster and provides safety information if the user feels anxious or scared based on the result of the user's emotion analysis.The server uses the emotion analysis result and the original text data from the fraudster as input, interrupts communication with the fraudster, and notifies the user with a warning message.The server obtains the warning message to be provided to the user as output.

[0887] 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.

[0888] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ) 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.

[0889] Another example of generative AI is Gemini (registered trademark) (Internet search engine). <url: https: gemini.google.com ?hl="ja">) are mentioned.

[0890] 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.

[0891] [Second embodiment]

[0892] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0893] 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.

[0894] 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).

[0895] 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.

[0896] 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.

[0897] 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).

[0898] 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.

[0899] 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.

[0900] 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.

[0901] 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.

[0902] 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.

[0903] Next, the specific processing by the specific processing unit 290 of the data processing device 12 will be described.

[0904] "Example 1"

[0905] One embodiment of the present invention provides a means for detecting unusual content in the speech techniques and documents of sophisticated fraudsters who access a user via telephone, email, etc. This means analyzes the speech techniques and documents of the fraudster and determines whether they contain fraudulent elements. For example, it can detect situations such as when a fraudster makes inappropriate requests to a user or when a fraudster provides information that misleads a user.

[0906] "Example 2"

[0907] Also provided is a means for suggesting responses to uncover the content of a fraudster. This means generates appropriate responses to the fraudster's speech or document content and provides them to the user. For example, if a fraudster makes an inappropriate request to the user, it is possible to suggest a response that ignores the request. Also, if a fraudster provides information that misleads the user, it is possible to suggest a response that denies that information.

[0908] "Example 3"

[0909] Furthermore, a means for communicating and responding to fraudsters is also provided. This means communicates with fraudsters on behalf of the user and reports the results to the user. For example, if a fraudster makes an inappropriate request of the user, the user can respond to the fraudster by ignoring the request. Also, if a fraudster provides information that misleads the user, the user can respond to the fraudster by denying that information.

[0910] "Example 4"

[0911] Finally, a method is provided to connect users only with content that is determined to be safe. This method provides content to users only if the scammer's speech or document content is determined to be safe as a result of interactions with the scammer. For example, if the scammer provides appropriate information to the user, that information can be provided to the user. Also, if the scammer makes inappropriate requests of the user or provides information that is misleading, such information will not be provided to the user.

[0912] The processing flow of each embodiment will be described below.

[0913] "Example 1"

[0914] Step 1: A method for detecting unusual content is activated in the speech techniques and documents of sophisticated fraudsters who access the victim via telephone, email, etc. This method analyzes the speech techniques and contents of documents of fraudsters and determines whether they contain any fraudulent elements.

[0915] Step 2: Generate appropriate responses to the scammer's speech and document content and provide them to the user. In this step, if the scammer makes an inappropriate request to the user, a response that ignores the request is suggested.

[0916] Step 3: Interact with the fraudster on behalf of the user and report the results to the user. In this step, if the fraudster makes an inappropriate request to the user, the system responds to the fraudster by ignoring the request.

[0917] Step 4: If the fraudster's language and document content are deemed safe after interaction with the fraudster, the content is provided to the user. In this step, if the fraudster has provided the user with appropriate information, the information is provided to the user.

[0918] "Example 2"

[0919] Step 1: A method for detecting unusual content is activated in the speech techniques and documents of sophisticated fraudsters who access the victim via telephone, email, etc. This method analyzes the speech techniques and contents of documents of fraudsters and determines whether they contain any fraudulent elements.

[0920] Step 2: Generate appropriate responses to the scammer's speech and document content and provide them to the user. In this step, if the scammer makes an inappropriate request to the user, a response that ignores the request is suggested.

[0921] Step 3: Interact with the fraudster on behalf of the user and report the results to the user. In this step, if the fraudster makes an inappropriate request to the user, the system responds to the fraudster by ignoring the request.

[0922] Step 4: If the fraudster's language and document content are deemed safe after interaction with the fraudster, the content is provided to the user. In this step, if the fraudster has provided the user with appropriate information, the information is provided to the user.

[0923] "Example 3"

[0924] Step 1: A method for detecting unusual content is activated in the speech techniques and documents of sophisticated fraudsters who access the victim via telephone, email, etc. This method analyzes the speech techniques and contents of documents of fraudsters and determines whether they contain any fraudulent elements.

[0925] Step 2: Generate appropriate responses to the scammer's speech and document content and provide them to the user. In this step, if the scammer makes an inappropriate request to the user, a response that ignores the request is suggested.

[0926] Step 3: Interact with the fraudster on behalf of the user and report the results to the user. In this step, if the fraudster makes an inappropriate request to the user, the system responds to the fraudster by ignoring the request.

[0927] Step 4: If the fraudster's language and document content are deemed safe after interaction with the fraudster, the content is provided to the user. In this step, if the fraudster has provided the user with appropriate information, the information is provided to the user.

[0928] "Example 4"

[0929] Step 1: A method for detecting unusual content is activated in the speech techniques and documents of sophisticated fraudsters who access the victim via telephone, email, etc. This method analyzes the speech techniques and contents of documents of fraudsters and determines whether they contain any fraudulent elements.

[0930] Step 2: Generate appropriate responses to the scammer's speech and document content and provide them to the user. In this step, if the scammer makes an inappropriate request to the user, a response that ignores the request is suggested.

[0931] Step 3: Interact with the fraudster on behalf of the user and report the results to the user. In this step, if the fraudster makes an inappropriate request to the user, the system responds to the fraudster by ignoring the request.

[0932] Step 4: If the fraudster's language and document content are deemed safe after interaction with the fraudster, the content is provided to the user. In this step, if the fraudster has provided the user with appropriate information, the information is provided to the user.

[0933] Example 1

[0934] Next, a description will be given of Example 1 of Form 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."

[0935] In modern society, there has been an increase in the number of sophisticated fraud methods carried out via telephone and email. This has increased the risk of many users becoming victims of fraud. Conventional methods have made it difficult to detect and respond to fraud, requiring users to rely on their own judgment to spot fraud. For this reason, there is a need for a system that automates fraud detection and response and protects users.

[0936] 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.

[0937] In this invention, the server includes a means for a user to input a message, a means for the terminal to send the message to the server, a means for the server to analyze the message, a means for the server to determine fraudulent elements using a generative AI model, a means for the server to generate an evaluation result, a means for the server to send the evaluation result to the terminal, and a means for the terminal to display the evaluation result to the user. This automates fraud detection and response, making it possible to reduce the risk of users becoming victims of fraud.

[0938] "Telephone and email" refers to a method of conveying information to users using communication means.

[0939] A "sophisticated fraudster" is someone who uses sophisticated verbal and written techniques to deceive users.

[0940] "Methods for detecting unusual content" are methods for finding unusual or unnatural elements in a message.

[0941] The "means for suggesting response content to detect fraudulent content" is a method for generating appropriate responses to potentially fraudulent messages.

[0942] "Methods of interacting with and responding to scammers" refers to methods for managing communications with scammers and responding appropriately.

[0943] "Means of connecting users only with messages that are deemed safe" is a method of providing users with only messages that are deemed to pose no risk of fraud.

[0944] "Means for users to input messages" refers to the method by which users input messages they receive into the system.

[0945] "Means for a terminal to send a message to a server" refers to a method by which a user's device sends an input message to a server.

[0946] The "means by which the server analyzes the message" refers to a method by which the server analyzes the message received using a natural language processing tool.

[0947] The "means by which the server uses the generative AI model to determine fraudulent elements" is the method by which the server uses the generative AI model to evaluate the fraudulent elements of a message.

[0948] The "means by which the server generates the evaluation result" refers to the method by which the server creates the evaluation result for the message based on the output of the generative AI model.

[0949] The "means by which the server transmits the evaluation results to the terminal" refers to a method by which the server transmits the evaluation results generated to the user's device.

[0950] The "means by which the terminal displays the evaluation results to the user" refers to a method by which the user's device displays the evaluation results it has received on its screen.

[0951] This invention is a system that protects users by detecting any unusual content in the speech techniques and documents of sophisticated fraudsters who access them via telephone, email, etc. This system is realized using a server, terminals, a generative AI model, and a natural language processing tool.

[0952] Hardware and software used

[0953] Hardware: Servers, devices (PCs, smartphones, etc.)

[0954] Software: Generative AI models (e.g., GPT-4), natural language processing tools (e.g., spaCy, NLTK)

[0955] Specific operation of the system

[0956] 1. The user types a message

[0957] Users input the contents of phone calls and emails they receive into the system. For example, if a user receives an email that they suspect is fraudulent, they can copy the contents of the email and paste it into the system's input field.

[0958] 2. The device sends a message to the server

[0959] The device (user's PC or smartphone) sends the entered message to the server. Specifically, it sends the message data to the server using an HTTP request.

[0960] 3. The server parses the message

[0961] The server analyzes the received message using natural language processing tools (e.g., spaCy or NLTK). Specifically, it analyzes the grammatical structure of the message and extracts important keywords and phrases.

[0962] 4. The server uses a generative AI model to determine fraudulent elements

[0963] The server uses a generative AI model (e.g., GPT-4) to determine whether the message content contains fraudulent elements. Specifically, the extracted keywords and phrases are incorporated into a prompt sentence and input into the generative AI model. For example, the prompt sentence "Does this message contain fraudulent elements?" is input into the generative AI model.

[0964] 5. The server generates the evaluation result

[0965] The server evaluates whether the message is suspicious based on the output of the generative AI model. Specifically, it analyzes the output of the generative AI model and generates an evaluation result such as "Possible fraud" or "Not likely fraud."

[0966] 6. The server sends the evaluation results to the device.

[0967] The server sends the generated evaluation results to the terminal. Specifically, it sends the evaluation results to the terminal using an HTTP response.

[0968] 7. The device displays the evaluation results to the user.

[0969] The device then displays the evaluation results to the user. Specifically, it displays a warning message on the user's PC or smartphone screen, such as "This message may be fraudulent."

[0970] Specific examples

[0971] Email example

[0972] Input message: "Congratulations! You are the winner of 1 million yen. Click this link for more details."

[0973] Prompt Sentence Examples

[0974] Prompt: "Does this message contain any elements of deception?"

[0975] Example output

[0976] Generative AI model output: "This message is fraudulent, especially since it asks you to click on a link."

[0977] Evaluation results

[0978] Server rating: "Possibly a scam"

[0979] Display results

[0980] Device displays: "This message may be fraudulent. Do not click on any links."

[0981] In this way, users can detect potentially fraudulent messages in advance and prevent damage before it occurs. This system automates fraud detection and response, reducing the risk of users becoming victims of fraud.

[0982] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0983] Step 1:

[0984] The user enters a message.

[0985] Input: The contents of phone calls and emails received by the user.

[0986] What happens: A user copies and pastes a suspected fraudulent message into an input field on the system.

[0987] Output: The input message data.

[0988] Step 2:

[0989] The device sends a message to the server.

[0990] Input: Message data entered by the user.

[0991] Specific operation: The device (user's PC or smartphone) sends message data to the server using an HTTP request.

[0992] Output: The message data sent to the server.

[0993] Step 3:

[0994] The server parses the message.

[0995] Input: The message data sent to the server.

[0996] What it does: The server uses natural language processing tools (e.g., spaCy or NLTK) to analyze the grammatical structure of the message and extract important keywords and phrases.

[0997] Output: Extracted keywords and phrases.

[0998] Step 4:

[0999] The server uses a generative AI model to determine fraudulent elements.

[1000] Input: Extracted keywords or phrases.

[1001] How it works: The server inputs a prompt sentence like "Does this message contain any elements of fraud?" into a generative AI model (e.g., GPT-4) and obtains the model's output.

[1002] Output: The output of the generative AI model (a rating of whether or not there is fraud).

[1003] Step 5:

[1004] The server generates the evaluation results.

[1005] Input: The output of a generative AI model.

[1006] Specific operation: The server analyzes the output of the generative AI model and generates an evaluation result such as "Possible fraud" or "Not likely to be fraudulent."

[1007] Output: The evaluation result.

[1008] Step 6:

[1009] The server transmits the evaluation results to the terminal.

[1010] Input: Evaluation result.

[1011] Specific operation: The server sends the evaluation results to the terminal using an HTTP response.

[1012] Output: Evaluation results sent to the device.

[1013] Step 7:

[1014] The terminal displays the evaluation results to the user.

[1015] Input: The evaluation result sent to the device.

[1016] Specific behavior: The device displays a warning message on the user's PC or smartphone screen, such as "This message may be fraudulent."

[1017] Output: The evaluation results displayed to the user.

[1018] (Application example 1)

[1019] Next, a description will be given of Application Example 1 of Form 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."

[1020] In recent years, sophisticated frauds using telephone and email have been increasing, increasing the risk of users becoming victims of fraud. In particular, it is difficult for users to detect frauds on their own when fraudsters use sophisticated speaking and writing techniques. Therefore, effective measures are needed to reduce the risk of fraud and ensure the safety of users.

[1021] 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.

[1022] In this invention, the server includes means for detecting anything unusual in the speech techniques or documents of sophisticated fraudsters who access the server by phone or email, means for suggesting responses to uncover the fraudster's content, means for communicating and responding to the fraudster, means for connecting only those determined to be safe to the user, means for analyzing voice data and detecting fraudulent elements, means for analyzing text data and detecting fraudulent elements, and means for issuing a warning if there is a possibility of fraud. This reduces the risk that users will be involved in fraud and ensures safe communication.

[1023] "Means for detecting unusual content" refers to a function that analyzes information received via telephone, email, etc., and automatically detects fraudulent elements or unnatural points.

[1024] The "means for suggesting response content to detect the content of a fraudster" is a function that generates appropriate responses to the fraudster's speech or writing and suggests them to the user.

[1025] "Means of interacting and responding to scammers" refers to functions for managing communication with scammers and taking appropriate action.

[1026] "Means of connecting users with only those things that are determined to be safe" is a function that provides users with only information and communications that are determined to pose no risk of fraud.

[1027] "Means for analyzing voice data and detecting fraudulent elements" refers to a function that analyzes voice data such as telephone conversations and detects elements that may be fraudulent.

[1028] "Means for analyzing text data and detecting fraudulent elements" refers to a function that analyzes text data such as emails and messages and detects elements that may be fraudulent.

[1029] "Means for issuing warnings in the event of possible fraud" is a function that issues a warning to the user when a risk of fraud is detected.

[1030] A system for implementing this invention includes means for detecting anything unusual in the speech techniques or documents of sophisticated fraudsters who access the system by phone or email, means for suggesting responses to uncover the fraudster's content, means for communicating and responding to fraudsters, means for connecting users only with those determined to be safe, means for analyzing voice data and detecting fraudulent elements, means for analyzing text data and detecting fraudulent elements, and means for issuing a warning if there is a possibility of fraud.

[1031] The server uses a speech recognition library to analyze the audio data and converts the audio file into text. The converted text is then analyzed by a generative AI model (e.g., a model using the transformers library) to determine whether it contains fraudulent elements. If fraudulent elements are detected, the server issues a warning to the user.

[1032] In addition, text data (e.g., emails and messages) is also analyzed by the generative AI model to determine whether it contains fraudulent elements. If fraudulent elements are detected, the server issues a warning to the user.

[1033] For example, if a user receives a message saying, "Please tell me your account information," the system will analyze the message and issue a warning if it determines there is a possibility of fraud. An example of a prompt sentence to input into the generative AI model is, "Please tell me your account information."

[1034] This system allows users to reduce the risk of being involved in fraud and ensure safe communication.

[1035] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1036] Step 1:

[1037] The server obtains the voice data or text data received from the user.

[1038] Input: Audio or text data

[1039] Output: Acquired audio data or text data

[1040] Step 2:

[1041] The server uses a speech recognition library to convert the voice data into text.

[1042] Input: Audio data

[1043] Output: Converted text data

[1044] Specific operation: Reads an audio file, performs speech recognition, and outputs it as text.

[1045] Step 3:

[1046] The server inputs the converted text data or directly received text data into the generative AI model.

[1047] Input: Text data

[1048] Output: Analysis results from the generative AI model

[1049] What it does: Text data is fed into a generative AI model (e.g., a model using the transformers library) and analyzed for deceptive elements.

[1050] Step 4:

[1051] The server determines whether the data contains any fraudulent elements based on the analysis results of the generative AI model.

[1052] Input: Analysis results of the generative AI model

[1053] Output: Judgment result on whether there is any fraudulent element

[1054] What it does: Evaluate the analysis results and flag any fraudulent elements detected.

[1055] Step 5:

[1056] The server will warn the user if any fraudulent elements are detected.

[1057] Input: Judgment result regarding the presence or absence of fraudulent elements

[1058] Output: Warning message

[1059] What it does: Sends a warning message to the user if there is a possibility of fraud.

[1060] Step 6:

[1061] If the server does not detect any fraudulent elements, it determines that the transaction is safe and proceeds with normal processing for the user.

[1062] Input: Judgment result regarding the presence or absence of fraudulent elements

[1063] Output: Normal processing result

[1064] What it does: Provide users with information or communications that are determined to be free of fraud risks.

[1065] Example 2

[1066] Next, a description will be given of Example 2 of Form 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."

[1067] In modern society, sophisticated fraud methods via telephone and email are on the rise. This increases the risk that many users will be deceived by the scammers' verbal tactics and documents, and lose their personal information and money. With conventional methods, it is difficult to properly analyze the scammers' messages and respond quickly and accurately, making it difficult to ensure the safety of users.

[1068] 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.

[1069] In this invention, the server includes means for receiving messages from fraudsters, means for transmitting the received messages to the server, means for inputting the messages received by the server into a generative AI model, means for the generative AI model to analyze the fraudster's message and generate an appropriate response content, means for transmitting the generated response content to the user's terminal, means for the user to check the suggested response content and reply to the fraudster as necessary, and means for connecting only those messages that are determined to be safe to the user. This makes it possible to respond quickly and accurately to fraudster's messages and ensure the safety of users.

[1070] A "fraudster" is someone who attempts to fraudulently obtain money or information by deceiving others.

[1071] A "message" is a communication sent by a fraudster to a user, and may be in the form of text, audio, images, etc.

[1072] A "server" is a computer system that processes and stores data over a network.

[1073] A "terminal" is a device that is directly operated by a user, and includes smartphones, tablets, personal computers, etc.

[1074] A "generative AI model" is an algorithm that uses artificial intelligence technology to generate and analyze text, such as a natural language processing model.

[1075] A "prompt" is an instruction entered into a generative AI model, which serves as a trigger for a specific response or analysis.

[1076] The "response content" is the message generated by the generative AI model as a result of analysis, and includes an appropriate response to the scammer.

[1077] "Analysis" is the process by which a generative AI model understands the input message and extracts meaning.

[1078] "Things that have been determined to be safe" refer to communication content and parties that have been determined to pose no risk of fraud.

[1079] "User" means an individual or organization that uses the system to respond to messages from scammers.

[1080] MODE FOR CARRYING OUT THE INVENTION

[1081] The present invention provides a system that generates appropriate responses to messages from fraudsters and helps users respond safely. A specific embodiment of this system is described below.

[1082] Hardware and software used

[1083] Hardware: Servers, terminals

[1084] Software: Generative AI models (e.g., GPT-4)

[1085] System Overview

[1086] The system receives messages from fraudsters, analyzes the messages, and generates appropriate responses. The generated responses are provided to the user, who can then use them to respond to the fraudster.

[1087] Specific actions

[1088] 1. The user receives a message from a scammer

[1089] A user receives a message from a scammer. For example, the scammer asks for your bank details.

[1090] 2. The device sends the received message to the server

[1091] The device sends the received message to the server. Specifically, the device application sends the message to the server via the API.

[1092] 3. The server inputs the received message into the generative AI model.

[1093] The server inputs the received message into the generative AI model. Specifically, the server passes the message to the generative AI model as a prompt. For example, the prompt might be, "I received a message from a scammer saying, 'Please tell me your bank account information.' Please suggest an appropriate response to this message."

[1094] 4. A generative AI model analyzes the scammer's message and generates an appropriate response.

[1095] The generative AI model analyzes the prompt and generates an appropriate response to the scammer's message, for example, "Sorry, but we can't give out any personal information. Let us know if there's anything else we can help you with."

[1096] 5. The server sends the generated response to the user's device.

[1097] The server sends the response received from the generated AI model to the user's device. Specifically, the server sends the response to the device via API.

[1098] 6. The user reviews the suggested responses and responds to the scammer if necessary

[1099] The user checks the suggested responses on their device. The user then uses the suggested responses to reply to the scammer. For example, the user might reply, "Sorry, but I can't give out any personal information."

[1100] Specific examples

[1101] For example, if a user receives the message "Please tell me your bank account information," the following prompt sentence is input into the generative AI model:

[1102] Example prompt sentence:

[1103] I received a message from a scammer asking me to provide my bank details. What would be a good response to this message?

[1104] Example response of the generative AI model:

[1105] "Sorry, I can't give out any personal information. Let me know if there's anything else I can help you with."

[1106] In this way, when a user receives a message from a scammer, the system uses a generative AI model to suggest an appropriate response, helping the user respond safely.

[1107] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1108] Step 1:

[1109] The user receives a message from the scammer.

[1110] Input: A message from the scammer (e.g., "Please tell me your bank details")

[1111] Output: Received message

[1112] What happens: The user's device receives and displays a message from the scammer.

[1113] Step 2:

[1114] The terminal sends the received message to the server.

[1115] Input: Received message

[1116] Output: Message sent to the server

[1117] Specific operation: The device application sends the received message to the server via API.

[1118] Step 3:

[1119] The server inputs the received message into the generative AI model.

[1120] Input: The message sent to the server

[1121] Output: The prompt input to the generative AI model

[1122] Specific operation: The server passes the received message to the generative AI model as a prompt. For example, the prompt might be, "I received a message from a scammer saying, 'Please tell me your bank account information.' Please suggest an appropriate response to this message."

[1123] Step 4:

[1124] A generative AI model analyzes the scammer's message and generates an appropriate response.

[1125] Input: prompt statement

[1126] Output: The generated response

[1127] How it works: The generative AI model analyzes the prompt and generates an appropriate response to the scammer's message. For example, the generative AI model might generate a response like, "Sorry, we can't give out any personal information. Let us know if there's anything else we can help you with."

[1128] Step 5:

[1129] The server sends the generated response to the user's terminal.

[1130] Input: Generated response content

[1131] Output: The response sent to the user's device

[1132] Specific operation: The server sends the response received from the generated AI model to the user's device via API.

[1133] Step 6:

[1134] The user reviews the suggested responses and responds to the scammer if desired.

[1135] Input: The response sent to the user's device

[1136] Output: Reply to the scammer

[1137] What happens: The user reviews the suggested responses on their device and responds to the scammer if necessary. For example, the user might reply, "Sorry, I can't give out my personal information."

[1138] (Application example 2)

[1139] Next, a description will be given of Application Example 2 of Form 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."

[1140] In recent years, sophisticated fraudulent activities using telephone and email have been increasing, increasing the risk of users being deceived by the scammers' verbal techniques and documents. There is a need for a method that allows users to respond appropriately to such fraudulent activities. There is also a need for a system that can detect signs of fraud in real time and provide appropriate responses.

[1141] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for detecting anything unusual in the speech techniques or documents of a clever fraudster who accesses the server by phone, email, etc., means for suggesting responses to uncover the fraudster's content, means for communicating and responding to the fraudster, means for connecting only those determined to be safe to the user, means for recording audio and converting the audio to text, means for determining the possibility of fraud using a generative AI model and generating an appropriate response, and means for displaying the generated response to the user. This allows users to reduce the risk of fraud and communicate safely.

[1142] "Telephone, email, etc." refers to means of communication such as voice calls and email.

[1143] A "sophisticated fraudster" is someone who uses sophisticated verbal and written techniques to deceive users.

[1144] "Means for detecting unusual content" refers to technology that automatically identifies unusual content, including indicators of fraud.

[1145] "Means for suggesting response content to detect fraudsters' content" refers to technology for generating appropriate responses to fraudsters' requests and information and providing them to users.

[1146] "Methods of interacting with and responding to fraudsters" refers to techniques used to manage communications with fraudsters and respond appropriately.

[1147] "Means of connecting users only with communications that are determined to be safe" refers to technology that provides users with only communications that are determined to be void of fraud risks.

[1148] "Means for recording audio and converting audio to text" refers to technology for recording audio data and converting that audio into text information.

[1149] "Means for using generative AI models to determine likelihood of fraud and generate appropriate responses" refers to technologies for using artificial intelligence models to assess likelihood of fraud and generate appropriate responses.

[1150] The term "means for displaying the generated response to the user" refers to a technique for visually presenting the generated response content to the user.

[1151] The system for implementing this invention has the function of detecting anything unusual in the speech techniques or documents of clever fraudsters who access the system by telephone or email, proposing responses to uncover the fraudster's content, communicating and responding to the fraudster, and connecting only those deemed safe to the user.

[1152] System configuration

[1153] The system consists of the following main components:

[1154] 1. Audio Recording Device: Use your smartphone's microphone to record audio.

[1155] 2. Speech Recognition Software: Converts recorded speech into text using a speech recognition library.

[1156] 3. Generative AI model: Uses OpenAI's API to analyze text data and determine the likelihood of fraud.

[1157] 4. Response Generator: Uses generative AI models to generate appropriate responses in cases of potential fraud.

[1158] 5. Display device: The smartphone display is used to display the generated response to the user.

[1159] Processing flow

[1160] 1. Voice recording and transcription: When a user answers a phone call, a voice recorder records the conversation, which is then converted into text by speech recognition software.

[1161] 2. Fraud Detection: A generative AI model analyzes text data to determine the likelihood of fraud, for example, using the following prompt:

[1162] Determine whether the following text is a scam and suggest an appropriate response:

[1163] Please tell us your bank account details

[1164] 3. Generate and display a response: If a fraudulent request is detected, the generative AI model generates an appropriate response and displays it to the user via a display device. For example, a response such as "Please ignore this request. It is extremely dangerous to give your bank account information to others" could be generated.

[1165] Specific examples

[1166] If a user receives the message "Please tell us your bank account information," the system operates as follows:

[1167] 1. A voice recorder records your conversation, which speech recognition software converts into text: "Please tell me your bank account details."

[1168] 2. A generative AI model analyzes this text and determines that it is likely to be fraudulent.

[1169] 3. The response generating device generates a response saying, "Please ignore this request. It is extremely dangerous to give your bank account information to others," and displays it to the user on the display device.

[1170] In this way, users can communicate safely and with less risk of fraud.

[1171] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1172] Step 1:

[1173] When the user answers the phone, the device's voice recorder records the conversation. The input is the audio of the conversation between the user and the fraudster, and the output is the recorded audio data, which is used in subsequent processing steps.

[1174] Step 2:

[1175] The device's speech recognition software (speech recognition library) converts the recorded voice data into text. The input is the recorded voice data, and the output is text data. This text data is used for fraud detection.

[1176] Step 3:

[1177] The server's generative AI model (OpenAI API) analyzes the text data and determines the likelihood of fraud. The input is the text data, and the output is the evaluation result regarding the likelihood of fraud. Specifically, the analysis is performed using the following prompt sentence:

[1178] Determine whether the following text is a scam and suggest an appropriate response:

[1179] Please tell us your bank account details

[1180] Step 4:

[1181] The server's response generator generates an appropriate response based on the evaluation results of the generative AI model. The input is the evaluation result regarding the possibility of fraud, and the output is the generated response text. For example, a response might be generated that reads, "Please ignore this request. It is very dangerous to give your bank account information to others."

[1182] Step 5:

[1183] The display device of the terminal displays the generated response text to the user. The input is the generated response text, and the output is the response information visually provided to the user. This allows users to communicate safely, reducing the risk of fraud.

[1184] Example 3

[1185] Next, a description will be given of Example 3 of Form Example 3. 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."

[1186] There is a problem that fraudsters try to deceive users by using clever speech techniques and documents over the phone, email, etc. Conventional systems have difficulty detecting fraudsters' fraudulent activities and responding appropriately, which increases the risk of users becoming victims of fraud. In addition, there is a lack of means to automate interactions with fraudsters and provide users with only safe information. This requires an effective way to prevent fraudsters' fraudulent activities while ensuring the safety of users.

[1187] The specific processing by the specific processing unit 290 of the data processing device 12 in the third embodiment is realized by the following means.

[1188] In this invention, the server includes means for receiving messages from fraudsters, means for analyzing the received messages, means for generating responses to the fraudsters using a generative AI model, means for sending the generated responses to the fraudsters, means for saving the results of interactions with the fraudsters, means for extracting safe information from the results of the interactions, and means for providing the safe information to users. This makes it possible to automatically detect fraudulent activities by fraudsters and respond appropriately. Furthermore, by providing only safe information to users, the safety of users can be ensured and fraud victims can be prevented.

[1189] A "fraudster" is someone who uses fraudulent means to deceive others and obtain money or information illegally.

[1190] A "message" is a communication sent by a scammer to a user, and may be in the form of text, audio, images, etc.

[1191] "Analysis" is the process of analyzing the content of received messages to determine their intent and whether or not they contain fraudulent activity.

[1192] A "generative AI model" is a system that uses artificial intelligence technology to generate and analyze text and speech, such as a natural language processing model.

[1193] A "response" is the reply that the generative AI model creates in response to a message from a scammer.

[1194] "Sending" is the act of sending the generated response to the fraudster over the network.

[1195] "Storage" refers to the act of recording the content of interactions with a fraudster in a storage device such as a database.

[1196] "Extraction" is the process of extracting specific information from the content of stored interactions.

[1197] "Safe information" is information that has been determined to pose no risk to users as a result of interactions with fraudsters.

[1198] "User" means an individual or organization using this system who is to be protected from fraudulent activity by fraudsters.

[1199] The invention is a system that receives messages from fraudsters, analyzes them, generates appropriate responses, and sends them to the fraudsters. It also aims to store the results of interactions with fraudsters, extract secure information, and provide it to users.

[1200] Hardware and software used

[1201] Hardware: Servers, terminals

[1202] Software: Generative AI models (e.g., GPT-4), database management systems (e.g., MySQL)

[1203] Specific operation of the system

[1204] Receiving messages from scammers

[1205] The server receives messages sent by fraudsters over the network. For example, a fraudster may send a message saying, "Please tell us your bank account information."

[1206] Message Parsing

[1207] The server inputs the received message into the generative AI model and analyzes its content. Specifically, the server inputs a prompt to the generative AI model: "Analyze the message from the scammer and determine whether it contains inappropriate requests or misleading information." The generative AI model then determines whether the message is inappropriate.

[1208] Generating a response

[1209] The server uses the generative AI model to generate an appropriate response to the fraudster. For example, the server inputs a prompt to the generative AI model saying, "Please generate and send an appropriate response to the fraudster." The generative AI model generates a response saying, "The request is inappropriate. We cannot provide personal information."

[1210] Sending a Response

[1211] The server sends the generated response to the fraudster. The server sends the response generated by the generative AI model to the fraudster over the network.

[1212] Saving the results of the interaction

[1213] The server stores the results of the interaction with the fraudster in a database. Specifically, the server stores the content of the interaction (messages from the fraudster, generated responses) in a database management system (e.g., MySQL).

[1214] Secure Information Extraction

[1215] The server analyzes the results of the interaction and extracts only the information that is deemed safe. The server inputs the content of the interaction stored in the database into the generative AI model again, and analyzes it using the prompt, "Analyze the results of the interaction with the scammer and extract only the information that is deemed safe."

[1216] Providing information to users

[1217] The server sends the information that has been determined to be safe to the user's terminal. The terminal displays the received information to the user. Specifically, the server sends the information that has been determined to be safe to the user's terminal via the network, and the terminal displays the received information on its screen and notifies the user.

[1218] Prompt Sentence Examples

[1219] "Analyze messages from scammers to determine if they contain inappropriate requests or misleading information."

[1220] "Generate and send an appropriate response to the scammer."

[1221] "Analyze the results of your interactions with the scammer and extract only the information that is deemed safe."

[1222] This system automates communication with fraudsters and provides only safe information to users, thereby preventing fraud damage. The flow of the identification process in the third embodiment will be described with reference to FIG.

[1223] Step 1:

[1224] Receiving messages from scammers

[1225] The server receives messages sent by fraudsters over the network.

[1226] Input: A message from the scammer (e.g., "Please tell me your bank details")

[1227] Output: Received message data

[1228] Specific operation: The server receives a message through the network interface and stores the message data in its internal memory.

[1229] Step 2:

[1230] Message Parsing

[1231] The server inputs the received message into a generative AI model and analyzes the content.

[1232] Input: Received message data

[1233] Output: Message analysis results (e.g., whether or not there are any inappropriate requests)

[1234] How it works: The server inputs message data into the generative AI model along with a prompt: "Analyze the message from the scammer and determine whether it contains inappropriate requests or misleading information." The generative AI model analyzes the content of the message and returns the results to the server.

[1235] Step 3:

[1236] Generating a response

[1237] The server uses a generative AI model to generate appropriate responses to the imposter.

[1238] Input: Message parsing result

[1239] Output: The generated response (e.g., "That request is inappropriate. We cannot give out personal information.")

[1240] Specific operation: The server inputs the prompt "Generate and send an appropriate response to the fraudster" and the analysis results to the generative AI model. The generative AI model generates an appropriate response and returns it to the server.

[1241] Step 4:

[1242] Sending a Response

[1243] The server generates a response and sends it to the impostor.

[1244] Input: The generated response

[1245] Output: Response transmission result (e.g. confirmation that the response was sent successfully)

[1246] Specific operation: The server sends the response generated by the generative AI model to the fraudster via the network, verifies the transmission result, and records it in a log.

[1247] Step 5:

[1248] Saving the results of the interaction

[1249] The server stores the results of the interaction with the fraudster in a database.

[1250] Input: Message from scammer, generated response, and submission result

[1251] Output: Save result (e.g., confirmation that the data was saved successfully to the database)

[1252] Specific operation: The server saves the contents of the exchange (message from the scammer, generated response, and transmission result) in a database management system (e.g., MySQL). It checks the saved results and records them in a log.

[1253] Step 6:

[1254] Secure Information Extraction

[1255] The server analyzes the results of the exchange and extracts only the information that is deemed safe.

[1256] Input: The conversation details stored in the database

[1257] Output: Secure information (e.g., information that can be provided to the user)

[1258] Specific operation: The server inputs the content of the interaction stored in the database into the generative AI model again, and analyzes it using the prompt, "Analyze the results of the interaction with the scammer and extract only the information that is determined to be safe." The generative AI model extracts the safe information and returns the results to the server.

[1259] Step 7:

[1260] Providing information to users

[1261] The server sends the information that has been determined to be safe to the user's terminal, which then displays the received information to the user.

[1262] Input: Secure Information

[1263] Output: Information displayed to the user

[1264] Specific operation: The server sends the information judged to be safe over the network to the user's device, which then displays the received information on its screen and notifies the user.

[1265] (Application example 3)

[1266] Next, a description will be given of Application Example 3 of Form Example 3. 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."

[1267] In recent years, there has been an increase in sophisticated fraud methods carried out via telephone and email, increasing the risk that users will be deceived by fraudsters' verbal tactics and documents. Elderly people and those with low digital literacy are particularly susceptible to fraud. In this situation, there is a need for a system that can automatically detect interactions with fraudsters and respond appropriately. It is also necessary to automatically generate appropriate responses to potentially fraudulent messages and provide users with only safe information.

[1268] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 3 is realized by the following means. In this invention, the server includes means for detecting anything unusual in the speech techniques or documents of a clever scammer who accesses the server by phone, email, etc., means for suggesting response content to uncover the scammer's content, means for interacting with and responding to the scammer, means for connecting the user to only those messages that are determined to be safe, means for automatically generating appropriate responses to messages that may be fraudulent, means for reporting the results of the interaction with the scammer to the user, and means for determining the possibility of fraud using a generative AI model and generating prompt text. This makes it possible to automatically detect interactions with scammers and respond appropriately.

[1269] The "means for detecting unusual content" is a function that automatically identifies unnatural expressions or requests that may be fraudulent in messages or call content received by telephone, email, etc.

[1270] The "means for proposing response content to detect the content of a fraudster" is a function for generating appropriate measures and responses to the information and requests provided by a fraudster.

[1271] "Means for interacting and responding to scammers" refers to a feature that automatically responds to messages and calls from scammers to reduce the risk of fraud.

[1272] "Means of connecting users with only those items that are deemed safe" is a function that notifies users of only information and requests that are deemed to pose no risk of fraud.

[1273] The "means for automatically generating an appropriate response to a potentially fraudulent message" is a function that automatically generates an appropriate response to a message that is suspected to be fraudulent.

[1274] "Means for reporting the results of interactions with scammers to users" is a function for reporting the content and results of interactions with scammers to users.

[1275] "Means for determining the likelihood of fraud using a generative AI model and generating prompt sentences" refers to a function that uses a generative AI model to assess the likelihood of fraud and create prompt sentences to generate appropriate responses.

[1276] The system for implementing this invention involves three parties working together: a server, a terminal, and a user. The server plays a central role in detecting interactions with fraudsters and generating appropriate responses. The terminal is a device through which the user receives messages and calls from fraudsters and communicates with the server. The user receives information provided by the system and takes action as necessary.

[1277] The server first uses a method to detect any unusual content in the speech techniques and documents of sophisticated fraudsters who access the server by phone, email, etc. This method uses natural language processing technology to analyze the contents of messages and calls and identify unnatural expressions and requests that may be fraudulent.

[1278] Next, a method is used to suggest responses to uncover fraudsters' content. This method uses a generative AI model to generate appropriate countermeasures and responses to the information and requests provided by fraudsters. The generative AI model has learned from past fraud cases and patterns, and is able to generate highly accurate responses.

[1279] Additionally, the system uses a means of communicating and responding to fraudsters, which automatically responds to messages and calls from fraudsters to mitigate the risk of fraud, for example by automatically generating appropriate responses to potentially fraudulent messages and responding to fraudsters.

[1280] In addition, a method is used to connect only those that are judged to be safe to the user. This method notifies the user only of information and requests that are judged to have no risk of fraud. This allows the user to receive only safe information.

[1281] The method also includes a means for reporting the results of interactions with fraudsters to the user. This means reports the content and results of interactions with fraudsters to the user, allowing the user to understand the situation.

[1282] For example, if a server receives a message from a fraudster saying, "Please tell us your bank account information," the server analyzes the message and determines that it is likely a scam. Using a generative AI model, the server automatically generates a response to the fraudster saying, "This request is inappropriate. Please ignore it." The user is then notified of the outcome of this interaction.

[1283] Examples of prompts include:

[1284] Message: "Please tell us your bank details."

[1285] Response: "This request is inappropriate. Please ignore it."

[1286] In this way, the server, terminal, and user can work together to reduce the risk of fraud and provide only safe information to the user.

[1287] The flow of the specific processing in Application Example 3 will be described with reference to FIG.

[1288] Step 1:

[1289] The server analyzes messages and call content received from the device. The input is the messages and call content received by the user. The server uses natural language processing technology to convert the messages and call content into text data and identify unnatural expressions and requests that may be fraudulent. The output is the analysis results.

[1290] Step 2:

[1291] The server determines whether there is a possibility of fraud based on the analysis results. The input is the analysis results obtained in step 1. The server detects keywords and patterns that may indicate fraud and assesses the risk of fraud. The output is a determination of whether there is a possibility of fraud.

[1292] Step 3:

[1293] If the server determines that there is a possibility of fraud, it uses a generative AI model to generate an appropriate response. The inputs are the fraud possibility determination result obtained in step 2, and the original message and call content. The server inputs a prompt sentence into the generative AI model to generate an appropriate response. The output is the generated response.

[1294] Step 4:

[1295] The server automatically sends the generated response to the impostor. The input is the response generated in step 3. The server sends an appropriate response to the impostor, mitigating the risk of fraud. The output is the response sent to the impostor.

[1296] Step 5:

[1297] The server notifies the user only of information or requests that are determined to be free of fraud risks. The input is the fraud possibility determination result obtained in step 2. The server notifies the user of information that is determined to be safe, allowing the user to receive only safe information. The output is the safe information that has been notified to the user.

[1298] Step 6:

[1299] The server reports the results of the interaction with the impostor to the user. The input is the content of the interaction with the impostor that took place in step 4. The server summarizes the results of the interaction and reports them to the user. The output is the results of the interaction reported to the user.

[1300] In this way, the server, terminal, and user can work together to reduce the risk of fraud and provide only safe information to the user.

[1301] 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.

[1302] "Example 1"

[1303] In one embodiment of the present invention, a system is provided in which a means for suggesting response content to detect fraudsters' messages is provided by generative AI, and further includes an emotion engine that recognizes the user's emotions. This system suggests response content to detect fraudsters' messages in response to information received by the user via telephone or email, and communicates and responds to fraudsters. Furthermore, the emotion engine analyzes emotions from the user's tone of voice and the text of emails, and provides the results to the system's means for communicating and responding to fraudsters.

[1304] "Example 2"

[1305] For example, if a user receives a call from a scammer, the system will detect something unusual about the scammer's speech and use generative AI to suggest a response that will expose the scammer's message. At the same time, the emotion engine will analyze the user's emotions from their tone of voice and provide the results to the system. The system will use this information to communicate and respond to the scammer.

[1306] "Example 3"

[1307] Additionally, if the emotion engine determines that the user's emotion is anxiety or fear, the system's means of interacting with and responding to the scammer will interrupt the interaction with the scammer and provide the user with safety information. For example, if the emotion engine determines that the user is feeling anxiety or fear in response to a call from a scammer, the system will automatically disconnect the call and notify the user that it was a call from a scammer and how to respond.

[1308] The processing flow of each embodiment will be described below.

[1309] "Example 1"

[1310] Step 1: The user receives information by phone or email.

[1311] Step 2: The system uses generative AI to suggest responses to identify the fraudster.

[1312] Step 3: The emotion engine analyzes the user's tone of voice and the content of the email to determine their emotions.

[1313] Step 4: The results of the emotion engine analysis are used to inform the system's interaction and response to fraudsters.

[1314] "Example 2"

[1315] Step 1: The user receives a call from a scammer.

[1316] Step 2: The system detects anything that seems odd about the scammer's speech and uses generative AI to suggest a response that will uncover the scammer's content.

[1317] Step 3: The emotion engine analyzes the emotion from the user's tone of voice and provides the results to the system.

[1318] Step 4: The system interacts with and responds to the scammer based on the results of the emotion engine analysis.

[1319] "Example 3"

[1320] Step 1: The emotion engine determines that the user's emotion is anxiety or fear.

[1321] Step 2: The system's means of interacting with and dealing with the scammers stops interacting with them.

[1322] Step 3: The system provides the user with secure information, such as informing them that the call was from a scammer and how to deal with it.

[1323] Example 1

[1324] Next, a description will be given of Example 1 of Form 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."

[1325] In modern society, sophisticated frauds using telephone and email are on the rise, making it difficult for users to detect these scams. Furthermore, users are unable to respond appropriately when interacting with fraudsters, which can lead to increased damage. Furthermore, while it is necessary to respond in a way that takes into account the user's feelings, current systems do not adequately address this issue. Effective methods to resolve these issues are needed.

[1326] 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.

[1327] In this invention, the server includes means for detecting inconsistent content in the speech techniques and documents of sophisticated fraudsters who access the server by telephone, email, etc., means for proposing response content to detect the fraudster's content, means for communicating and responding to the fraudster, means for connecting only those determined to be safe to the user, means for analyzing the user's emotions, and means for adjusting the response content based on the analyzed emotions. This makes it possible to quickly detect fraudulent content and propose appropriate responses to the user. Furthermore, by providing response content that takes the user's emotions into consideration, an environment in which the user can respond with peace of mind can be provided.

[1328] "Telephone and email" refers to the means of sending and receiving information using voice communication and text messages.

[1329] A "sophisticated fraudster" is someone who commits fraud using sophisticated verbal and written techniques to deceive users.

[1330] "Means for detecting unusual content" refers to technology that automatically identifies fraudulent elements or unnatural information.

[1331] The "means for suggesting response content to detect fraudsters" is a technology that generates an appropriate response to detect fraudulent activity.

[1332] "Methods of interacting and responding to fraudsters" refers to techniques for interacting with fraudsters and taking appropriate action.

[1333] "Means of connecting users with only those things that are determined to be safe" is a technology that provides users with only information and communications that are determined to pose no risk of fraud.

[1334] "Means for analyzing user emotions" refers to technology that identifies emotions from the tone of a user's voice and the content of their text.

[1335] "Means for adjusting response content based on analyzed emotions" refers to technology that optimizes response content according to the user's emotional state.

[1336] "Generative AI" is an AI technology that uses natural language processing and machine learning to generate text and speech.

[1337] The present invention provides a system that helps users respond appropriately to sophisticated fraudulent activities using telephone and email. The system includes multiple means for detecting fraudulent content and suggesting appropriate responses.

[1338] Hardware and software used

[1339] 1. Server

[1340] Speech Recognition Software: Uses the Google Cloud Speech-to-Text API to convert received voice data into text.

[1341] Natural language processing software: spaCy is used to analyze text data and extract grammar and meaning.

[1342] Generative AI model: OpenAI GPT-4 is used to detect fraudulent elements and generate appropriate response content.

[1343] Sentiment analysis engine: Uses IBM Watson Tone Analyzer to analyze user emotions.

[1344] 2. Terminal

[1345] Voice recording function: Records the voice of the call and sends it to the server.

[1346] Text sending function: Sends the contents of received emails to the server.

[1347] Display function: Presents the generated response content to the user.

[1348] 3. Users

[1349] Receiving phone calls or emails: You may receive contact from scammers.

[1350] Review responses: Review the responses provided and communicate with the scammer.

[1351] Specific operation of the system

[1352] 1. Receiving Data

[1353] The user receives a phone call or email.

[1354] The terminal transmits the received voice data and text data to the server.

[1355] 2. Data Analysis

[1356] The server parses the received data.

[1357] For audio data, the server converts the audio to text using the Google Cloud Speech-to-Text API.

[1358] For text data, the server uses spaCy to parse the document and extract grammar and semantics.

[1359] 3. Fraud detection

[1360] The text data analyzed by the server is input into a generative AI model (OpenAI GPT-4) to determine whether it contains any fraudulent elements.

[1361] The server detects certain keywords and phrases (e.g., "bank account," "fraud") and assesses the likelihood of fraud.

[1362] 4. Emotion Analysis

[1363] The device inputs the user's tone of voice and the text of the email into an emotion engine (IBM Watson Tone Analyzer) to analyze the user's emotions.

[1364] The device sends the results of the emotion analysis to the server.

[1365] 5. Response suggestions

[1366] The server uses a generative AI model to generate appropriate response content to detect the fraudster's content.

[1367] The server takes into account the results of the emotion analysis and adjusts the response content to make the user feel at ease.

[1368] 6. Providing a Response

[1369] The terminal presents the generated response to the user.

[1370] The user uses the provided responses to interact with the fraudster.

[1371] Specific examples

[1372] For example, if a user receives an email saying, "Your bank account has been fraudulently used. Please contact us immediately," the system operates as follows:

[1373] 1. Receiving Data

[1374] The user receives the email and the terminal sends the contents to the server.

[1375] 2. Data Analysis

[1376] The server analyzes the text of the email and detects any fraudulent elements.

[1377] 3. Fraud detection

[1378] The server detects the phrase "fraudulent use of bank account" and determines that it is likely a scam.

[1379] 4. Emotion Analysis

[1380] The device analyzes the user's emotions and transmits to the server that the user is feeling anxious.

[1381] 5. Response suggestions

[1382] The server uses the generative AI model to generate a response that reads, "This email may be fraudulent. We recommend that you contact the bank directly through their official website."

[1383] 6. Providing a Response

[1384] The terminal presents the generated response content to the user, and the user acts based on the content.

[1385] Prompt Sentence Examples

[1386] An example of a prompt sentence to be input to the generative AI model is as follows.

[1387] I received an email saying "Your bank account has been compromised. Please contact us immediately." Please help me determine if this email is a scam and suggest an appropriate response.

[1388] In this way, the system helps users respond appropriately to the sophisticated verbal and written techniques of fraudsters.

[1389] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1390] Step 1:

[1391] Receiving data

[1392] The user receives a phone call or email.

[1393] The terminal transmits the received voice data and text data to the server.

[1394] Input: Telephone voice data, email text data.

[1395] Output: Audio and text data sent to the server.

[1396] Specific operation: The device records the audio of the phone call and sends the contents of the email directly to the server.

[1397] Step 2:

[1398] Data analysis

[1399] The server parses the received data.

[1400] For audio data, the server converts the audio to text using the Google Cloud Speech-to-Text API.

[1401] For text data, the server uses spaCy to parse the document and extract grammar and semantics.

[1402] Input: Audio and text data sent to the server.

[1403] Output: Parsed text data.

[1404] Specific operation: The server converts the voice data into text and passes the text to a natural language processing engine for analysis.

[1405] Step 3:

[1406] Fraud detection

[1407] The text data analyzed by the server is input into a generative AI model (OpenAI GPT-4) to determine whether it contains any fraudulent elements.

[1408] The server detects certain keywords and phrases (e.g., "bank account," "fraud") and assesses the likelihood of fraud.

[1409] Input: Parsed text data.

[1410] Output: A score for whether or not the fraud is likely.

[1411] How it works: The server inputs text data into a generative AI model and receives a score indicating whether it is likely to be fraudulent.

[1412] Step 4:

[1413] Emotion Analysis

[1414] The device inputs the user's tone of voice and the text of the email into an emotion engine (IBM Watson Tone Analyzer) to analyze the user's emotions.

[1415] The device sends the results of the emotion analysis to the server.

[1416] Input: User's tone of voice, email text.

[1417] Output: Parsed emotion data.

[1418] Specific operation: The device passes voice and text data to the emotion engine and sends the analysis results to the server.

[1419] Step 5:

[1420] Response suggestions

[1421] The server uses a generative AI model to generate appropriate response content to detect the fraudster's content.

[1422] The server takes into account the results of the emotion analysis and adjusts the response content to make the user feel at ease.

[1423] Input: Fraud likelihood score, parsed sentiment data.

[1424] Output: The generated response.

[1425] Specific operation: The server inputs a prompt sentence into the generative AI model and fine-tunes the generated response content based on the results of emotion analysis.

[1426] Step 6:

[1427] Providing a response

[1428] The terminal presents the generated response to the user.

[1429] The user uses the provided responses to interact with the fraudster.

[1430] Input: The generated response content.

[1431] Output: The response presented to the user.

[1432] Specific operation: The device displays the response content on the screen, and the user confirms the content.

[1433] (Application example 1)

[1434] Next, a description will be given of Application Example 1 of Form 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."

[1435] There is an increasing risk that users will easily be deceived by the persuasive tactics and written messages of scammers who contact them cleverly via phone or email. Particularly problematic are the clever tactics used by scammers that lead users to respond to inappropriate requests or be misled by misleading information. Furthermore, users often feel anxiety and stress about scams, and appropriate responses to these issues are required. To solve these issues, a system is needed that automates fraud detection and response, and analyzes users' emotions to provide appropriate support.

[1436] 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.

[1437] In this invention, the server includes means for detecting anything unusual in the speech techniques or documents of clever scammers who access the server by phone or email, means for suggesting responses to uncover the scammer's content, means for communicating and responding to the scammer, means for analyzing the user's emotions, and means for connecting the user only with connections that are determined to be safe. This reduces the risk of fraud and allows users to communicate with peace of mind.

[1438] "Means for detecting unusual content" refers to technology that detects unnatural or inappropriate content in the speech techniques or documents of sophisticated fraudsters who access you via phone or email.

[1439] The "means for suggesting response content to detect fraudsters' content" is a technology for generating and suggesting appropriate responses to inappropriate requests or misleading information that fraudsters make to users.

[1440] "Means for communicating and responding to fraudsters" refers to technology that automatically responds to contact from fraudsters and takes appropriate action.

[1441] "Means for analyzing user emotions" refers to technology that analyzes emotions from the tone of a user's voice or the text of an email, and provides the results to other functions of the system.

[1442] "Means of connecting users only with communications that are determined to be safe" is a technology that provides users with only communications that are determined to be void of fraud risks.

[1443] "Generative AI" is an artificial intelligence technology that automatically generates responses to detect fraudsters' messages.

[1444] An "emotion engine" is a technology that analyzes the user's emotions and provides the results to other functions of the system.

[1445] As an embodiment of this invention, we will explain how to install a fraud detection assistant system on a smartphone. This system has the functions of detecting anything unusual in the speech techniques and documents of sophisticated fraudsters who access the smartphone via phone or email, suggesting responses to uncover the fraudster's content, and analyzing the user's emotions.

[1446] Hardware and software used

[1447] Hardware: Smartphone

[1448] Software: Python, regular expression library, sentiment analysis library, generative AI library

[1449] System configuration

[1450] 1. How to detect inappropriate content:

[1451] The server uses a regular expression library to analyze the content of calls and emails to detect whether they contain fraudulent patterns, thereby identifying messages that pose a risk of fraud.

[1452] 2. Suggested response methods to detect scammers:

[1453] The server uses a generative AI library to generate appropriate responses to the scammer's content, providing guidance to users on how to respond appropriately to the scammer.

[1454] 3. How to communicate and respond to scammers:

[1455] The server automatically responds to contact from fraudsters and takes appropriate action, allowing users to safely communicate with fraudsters.

[1456] 4. Means of analyzing user sentiment:

[1457] The server uses an emotion analysis library to analyze the emotion of the user's tone of voice and the content of the email, allowing it to provide additional support if the user is feeling stressed or anxious.

[1458] 5. Methods to connect users only with devices that are deemed safe:

[1459] The server will only provide users with communications that are deemed to be fraud-free, allowing users to communicate with peace of mind.

[1460] Specific examples

[1461] For example, if a user receives a message saying, "Please provide your bank account information. Urgent action is required," the system will operate as follows:

[1462] 1. Detecting unusual content:

[1463] The server uses a regular expression library to parse the message and detect that the phrase "Please tell me your bank account details" matches a fraudulent pattern.

[1464] 2. Response suggestions:

[1465] The server uses a generative AI library to generate an appropriate response: "Sorry, we cannot provide any personal information."

[1466] 3. Emotion analysis:

[1467] The server uses an emotion analysis library to analyze whether the user is feeling stressed by this message.

[1468] 4. Providing secure communications:

[1469] The server determines that this message poses a risk of fraud and displays a warning to the user.

[1470] Prompt Sentence Examples

[1471] message = "Please provide your bank details. This is urgent."

[1472] result = assistant.handle_message(message)

[1473] print(result)

[1474] In this way, the fraud detection assistant system helps users reduce the risk of being scammed and enables them to communicate with peace of mind.

[1475] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1476] Step 1:

[1477] The server receives phone and email messages. As input, it provides the message received by the user. The server uses a regular expression library to parse this message. Specifically, it scans the text in the message for phrases that match fraud patterns. As output, it generates a flag indicating whether the message is suspected of fraud.

[1478] Step 2:

[1479] If the server determines that fraud is suspected, it uses a generative AI library to generate an appropriate response. The server receives a suspected fraudulent message as input. Based on this message, the server generates an appropriate response to the fraudster. The generated response is the output.

[1480] Step 3:

[1481] The server uses a sentiment analysis library to analyze the user's sentiment. As input, it receives a message from the user. The server analyzes the text and audio data of the message to assess the level of stress and anxiety felt by the user. As output, it obtains the results of the sentiment analysis.

[1482] Step 4:

[1483] The server displays a warning to the user based on the suspected fraudulent message and the results of sentiment analysis. As input, the server receives the suspected fraudulent message and the results of sentiment analysis. Based on this information, the server notifies the user of the risk of fraud and prompts them to take appropriate action. As output, a warning message is generated that is displayed to the user.

[1484] Step 5:

[1485] The server provides only messages that are determined to be safe to the user. As input, messages that are determined to be free of fraud are provided. The server forwards these messages to the user, allowing them to communicate with confidence. As output, safe messages are provided to the user.

[1486] In this way, the fraud detection assistant system helps users reduce the risk of being scammed and enables them to communicate with peace of mind.

[1487] Example 2

[1488] Next, a description will be given of Example 2 of Form 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."

[1489] Conventional systems have difficulty responding appropriately to the sophisticated speech techniques and documents of fraudsters, putting users at high risk of being scammed. Furthermore, they are unable to take into consideration the user's feelings when interacting with fraudsters, often resulting in stress and anxiety. This has led to a need for effective measures to ensure user safety.

[1490] 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.

[1491] In this invention, the server includes a means for detecting any inconsistencies in the speech techniques and documents of sophisticated scammers who access the server by phone, email, etc., a means for proposing response content to uncover the scammer's content, a means for communicating and responding to the scammer, a means for connecting only those determined to be safe to the user, a means for analyzing emotions from the user's tone of voice, and a means for proposing an optimal response based on the generated response content and the emotion analysis results. This makes it possible to respond appropriately to the scammer's clever speech techniques and documents and ensure the user's safety. Furthermore, by taking the user's emotions into consideration when responding, stress and anxiety can be reduced.

[1492] "Telephone, email, etc." refers to means of communication such as voice calls and email.

[1493] A "fraudster" is someone who attempts to fraudulently obtain money or information by deceiving others.

[1494] "Speaking" refers to the skill and techniques of using words skillfully.

[1495] "Document" refers to written information or messages.

[1496] "Unusual content" refers to unnatural and suspicious content that differs from normal communication.

[1497] "Means for sensing" refers to a method or device for identifying and detecting a particular condition or pattern.

[1498] "Response content" refers to the content of a reply to the other person's words or actions.

[1499] "Suggestion means" refers to a method or device for presenting a particular action or option.

[1500] "Means of communication and response" refers to the methods and devices used to communicate and respond to others.

[1501] "Things that have been determined to be safe" refers to information or communications that are deemed to be low risk and safe for users.

[1502] "User" refers to any individual or entity that uses the System.

[1503] "Tone of voice" refers to the pitch, intensity, and emotional expression of the voice.

[1504] "Means for analyzing emotions" refers to methods and devices for identifying and analyzing emotions from voice, facial expressions, etc.

[1505] "Generated response content" refers to a response content automatically generated by the system.

[1506] "Emotion analysis results" refers to information obtained as a result of analyzing emotions.

[1507] The "means for suggesting an optimal response" refers to a method or device for suggesting the most appropriate response depending on the situation.

[1508] "Generative AI" refers to AI that has the ability to generate new information and responses based on data.

[1509] "Voice and email services" refers to communications services that provide voice calls and email.

[1510] "Option Menu" refers to selectable functions and services provided in addition to the basic functions.

[1511] The present invention provides a system for ensuring the safety of users by appropriately dealing with the clever speech techniques and documents of fraudsters. A specific embodiment of this system will be described below.

[1512] Hardware and software used

[1513] Hardware: Servers, devices (smartphones and PCs)

[1514] Software: Generative AI models (e.g., GPT-4), emotion engines

[1515] System Overview

[1516] The system analyzes the scammer's speaking style and the content of the document, generates appropriate responses, and provides them to the user. Specifically, it suggests ignoring or denying responses to inappropriate requests or misleading information from the scammer.

[1517] Data processing and calculation flow

[1518] 1. The user receives a call from a scammer.

[1519] 2. The device records the call and sends it to the server in real time.

[1520] 3. The server receives the recording data and inputs it into the generative AI model.

[1521] 4. The generative AI model detects anything that seems odd about the scammer's speech and generates an appropriate response.

[1522] 5. The emotion engine analyzes the emotion from the tone of the user's voice and provides the results to the server.

[1523] 6. The server proposes the optimal response to the user based on the generated response content and the results of sentiment analysis.

[1524] Specific examples

[1525] Scenario: A user is asked by a scammer to "provide your bank account details."

[1526] An example prompt for a generative AI model: "If a scammer is asking for your bank account information, suggest how you should respond."

[1527] Implementation details

[1528] When a user receives a call from a scammer, the device records the conversation and sends it to a server in real time.

[1529] The server receives the recorded data and inputs it into a generative AI model (e.g., GPT-4), which detects anything inconsistent with the scammer's speech and generates a response such as, "The request is inappropriate. Please do not provide any personal information."

[1530] At the same time, the emotion engine analyzes the user's emotions from the tone of their voice and provides the server with results such as "the user is feeling anxious."

[1531] The server then proposes the best response to the user based on the generated response and the results of sentiment analysis, providing specific advice such as "Please ignore the request. Do not provide personal information."

[1532] In this way, the system can detect the fraudster's behavior and suggest appropriate responses to the user, ensuring user safety and preventing fraud before it occurs. It also takes into account the user's emotions, reducing stress and anxiety.

[1533] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1534] Step 1:

[1535] A user receives a call from a scammer.

[1536] What happens: The user receives a call on their smartphone in which a scammer asks for their bank account details.

[1537] Enter: A call from a scammer.

[1538] Output: Audio data of the call.

[1539] Step 2:

[1540] The device records the call and sends it to the server in real time.

[1541] Specific operation: The smartphone's recording function is activated, and the call is recorded while streaming the data to the server.

[1542] Input: Audio data from the call.

[1543] Output: The audio data sent to the server.

[1544] Step 3:

[1545] The server inputs the recorded data into a generative AI model.

[1546] How it works: The server analyzes the recording, converts the scammer's speech into text data, and feeds it into a generative AI model (e.g., GPT-4).

[1547] Input: The audio data sent to the server.

[1548] Output: The text data fed into the generative AI model.

[1549] Step 4:

[1550] The generative AI model generates appropriate response content.

[1551] Specific behavior: The generative AI model generates a response such as, "That request is inappropriate. Please do not provide personal information."

[1552] Input: Text data fed into the generative AI model.

[1553] Output: The generated response.

[1554] Step 5:

[1555] The emotion engine analyzes emotions from the user's tone of voice.

[1556] Specific operation: The emotion engine analyzes the tone of the user's voice and sends the emotion analysis result, such as "the user is feeling anxious," to the server.

[1557] Input: Audio data from the call.

[1558] Output: Emotion analysis results.

[1559] Step 6:

[1560] The server proposes the optimal response to the user based on the generated response content and the results of sentiment analysis.

[1561] Specific action: The server notifies the user by sending specific advice to the user's device, such as "Please ignore the request. Do not provide personal information."

[1562] Input: Generated response content, sentiment analysis results.

[1563] Output: The best response suggested to the user.

[1564] (Application example 2)

[1565] Next, a description will be given of Application Example 2 of Form 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."

[1566] In recent years, sophisticated fraudulent activities using telephone and email have been on the rise, increasing the risk that users will be deceived by the scammers' verbal tactics and written documents. Furthermore, delayed responses to fraudulent activities can increase users' mental stress and anxiety. Conventional systems struggle to detect fraudulent activities in real time and generate appropriate responses, and are also inadequate in responding while taking into account the user's emotional state. Therefore, there is a need for a system that can reduce the mental burden on users while ensuring their safety.

[1567] 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.

[1568] In this invention, the server includes a means for detecting anything unusual in the speech techniques or documents of sophisticated fraudsters who access the server by phone or email, a means for suggesting responses to uncover the fraudster's content, a means for communicating and responding to the fraudster, a means for analyzing the user's emotional state and providing advice to reduce stress and anxiety, and a means for connecting the user only to those connections that are determined to be safe. This makes it possible to detect fraudulent activity in real time and generate appropriate responses. It also makes it possible to respond in consideration of the user's emotional state, thereby reducing mental burden.

[1569] "Telephone, email, etc." refers to means of communication between users and the outside world, including voice communication and email.

[1570] A "sophisticated fraudster" is someone who commits fraud using sophisticated verbal and written techniques to deceive users.

[1571] "Perlocution and writing" refers to the words and written content that scammers use to deceive users.

[1572] "Methods for detecting unusual content" are techniques for identifying unnatural or suspicious elements in a fraudster's speech or documents.

[1573] The "means for suggesting response content to detect the content of a fraudster" is a technology for generating appropriate responses to the speech or writing of a fraudster and providing them to the user.

[1574] "Methods of interacting with and responding to fraudsters" are techniques for managing communication with fraudsters and responding appropriately.

[1575] "Means for analyzing the user's emotional state and providing advice to reduce stress and anxiety" refers to technology that analyzes the user's emotions and provides advice to reduce mental burden.

[1576] "Means of connecting users only with communications that are determined to be safe" is a technology that connects users only with communications that are determined to pose no risk of fraud.

[1577] "Generative AI" is a technology that uses artificial intelligence to generate text and speech.

[1578] "Emotion analysis" is a technology that identifies emotions from a user's tone of voice and text.

[1579] The system for implementing this invention detects any inconsistencies in the speech techniques or documents of sophisticated scammers who access the user via phone or email, and suggests responses to uncover the scammer's content. Furthermore, the system interacts with and responds to the scammer, analyzes the user's emotional state, and provides advice to reduce stress and anxiety. It also has the function of connecting the user only to those connections that are determined to be safe.

[1580] Hardware and software used

[1581] Hardware: Smartphone (microphone, speaker)

[1582] software:

[1583] Python

[1584] Speech Recognition Library (Speech Recognition)

[1585] Transformers library (generative AI models)

[1586] TextBlob library (sentiment analysis)

[1587] System processing overview

[1588] 1. Speech recognition: The device (smartphone) uses a microphone to record voice and converts the voice into text using a speech recognition library.

[1589] 2. Fraud detection: The server checks whether the converted text contains certain keywords (e.g., bank account, password).

[1590] 3. Sentiment Analysis: The server analyzes the sentiment of the text using the TextBlob library. If the sentiment is negative, it determines that the likelihood of fraud is high.

[1591] 4. Response generation: The server uses a generative AI model (GPT-3) to generate an appropriate response.

[1592] 5. Displaying the results: The terminal displays the generated response to the user.

[1593] Specific examples

[1594] If a user receives a call from a scammer asking for their bank account information, the system will detect this and generate a response saying, "This may be a scam. Please ignore."

[1595] Prompt Sentence Examples

[1596] When receiving a call from a scammer, the following prompt is fed into the generative AI model:

[1597] "This may be a scam. Please ignore it."

[1598] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1599] Step 1:

[1600] The device (smartphone) uses a microphone to record the user's voice. The recorded voice data becomes the input. The device converts this voice data into text using a speech recognition library. The converted text becomes the output.

[1601] Step 2:

[1602] The server receives the text obtained in step 1 and performs fraud detection. Specifically, it checks whether the text contains specific keywords (e.g., bank account, password). If any of the keywords are detected, the text is determined to be potentially fraudulent. This determination result is output.

[1603] Step 3:

[1604] The server performs sentiment analysis on the text determined to be potentially fraudulent in step 2 using the TextBlob library. The input is the text from step 2, and the analysis results are a score of positive or negative sentiment. If the sentiment is negative, it is determined to be highly likely to be fraudulent.

[1605] Step 4:

[1606] The server generates an appropriate response using a generative AI model (GPT-3) based on the emotion analysis results from step 3. The input is the emotion analysis results from step 3 and text, and the generated response is the output. For example, a response such as "This may be a scam. Please ignore." is generated.

[1607] Step 5:

[1608] The terminal displays the response content generated in step 4 to the user. The input is the response content of step 4, which is presented to the user visually or audibly. The user can confirm this response content and take appropriate action.

[1609] Example 3

[1610] Next, a description will be given of Example 3 of Form Example 3. 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."

[1611] Fraudulent acts by fraudsters using sophisticated speech or documents can cause serious damage to users. Conventional systems have had difficulty detecting fraudulent acts by fraudsters and responding appropriately. In addition, there was a lack of means to respond quickly when users felt anxious or scared. This made it difficult to ensure the safety of users.

[1612] The specific processing by the specific processing unit 290 of the data processing device 12 in the third embodiment is realized by the following means.

[1613] In this invention, the server includes means for receiving communications from fraudsters, means for analyzing the messages and determining the emotional state, means for generating a response to the fraudster, means for sending the response and reporting the results, means for providing only safe information to the user, and means for interrupting the interaction via the emotion engine. This allows fraudulent activity by fraudsters to be detected quickly and dealt with appropriately. Furthermore, if the user feels anxious or scared, a quick response can be made to ensure the user's safety.

[1614] "Means for receiving communications from scammers" refers to functions for receiving communications such as phone calls, emails, and messages from scammers.

[1615] The "means for analyzing messages and determining emotional state" is a function for analyzing received messages and determining the user's emotional state (anxiety, fear, etc.).

[1616] The "means for generating a response to an imposter" is a function for generating an appropriate response to an imposter.

[1617] The "means for sending a response and reporting the result" is a function for sending the generated response to the fraudster and reporting the result to the user.

[1618] "Means of providing only safe information to users" is a function that analyzes the content of interactions with scammers and provides users with only information that is deemed safe.

[1619] "Means to interrupt interaction using emotion engine" is a function that interrupts interaction with a scammer if the user's emotions are judged to be anxiety or fear.

[1620] A "generative AI model" is an artificial intelligence model used to generate responses to fraudsters.

[1621] This invention is a system for receiving communications from fraudsters and responding appropriately. The system is implemented using a server, terminals (PCs and smartphones), a generative AI model, an emotion engine, and a communication module.

[1622] Hardware and software used

[1623] Hardware: Servers, terminals (PCs, smartphones)

[1624] Software: Generative AI models (e.g., GPT-4), emotion engines, and communication modules

[1625] Specific operation of the system

[1626] 1. Receiving communications from scammers

[1627] The server receives communications (phone calls, emails, messages, etc.) from the fraudster.

[1628] The server logs the communications it receives.

[1629] Example: When a server receives a call from a scammer, it records the call and stores it in a database.

[1630] 2. Message analysis and emotional state determination

[1631] The server sends the data to the emotion engine for analysis of the received message.

[1632] The emotion engine analyzes the content of the message and determines the user's emotional state (anxiety, fear, etc.).

[1633] Example: The server sends a message from a scammer to the emotion engine, which determines that the user is feeling anxious.

[1634] 3. Generating responses to impostors

[1635] The server sends prompts to the generative AI model, which generates appropriate responses to the imposter.

[1636] The generative AI model generates a response based on the prompt sentence and sends it back to the server.

[1637] Example: The server sends the generative AI model a prompt saying, "Generate an appropriate response to a message from a scammer. Message content: 'Please tell us your bank account information.'" The generative AI model then generates a response saying, "We are unable to provide such information."

[1638] 4. Sending Responses and Reporting Results

[1639] The server sends the generated response to the impostor.

[1640] The server reports the results of the interaction to the user.

[1641] Example: The server generates a response and sends it to the fraudster, then notifies the user that it has responded appropriately to the fraudster's request.

[1642] 5. Providing safe information

[1643] The server analyzes the content of the exchange with the scammer and determines whether it is safe.

[1644] The server provides users with only information that is deemed safe.

[1645] Example: If the server analyzes a message from a scammer and determines that the information is safe, it notifies the user of that information.

[1646] 6. Emotional Engine Interrupts Interactions

[1647] If the emotion engine determines that the user's emotions are anxiety or fear, it will discontinue the interaction with the scammer.

[1648] The server automatically disconnects the call with the scammer and notifies the user that the call was from a scammer and how to deal with the situation.

[1649] Example: If the emotion engine determines that the user is feeling scared, the server will disconnect the call and notify the user that the call was from a scammer and that they should not answer calls from this number in the future.

[1650] Prompt Sentence Examples

[1651] "Generate an appropriate response to the scammer's message: 'Please provide your bank details.'"

[1652] "Please analyze the results of your interactions with scammers and only provide safe information to your users."

[1653] "Generate ways for users to respond if they feel anxious or scared."

[1654] This system provides advanced technology to automate interactions with fraudsters and ensure user safety. The flow of the identification process in the third embodiment will be described with reference to FIG.

[1655] Step 1:

[1656] Receiving communications from scammers

[1657] The server receives communications (phone calls, emails, messages, etc.) from the fraudster.

[1658] Input: Communication data from the scammer (voice, text, etc.)

[1659] Data processing: Log communication data and convert voice data to text as needed.

[1660] Output: Communication data in a parsable format

[1661] What it does: When the server receives a call from a scammer, it records the call, converts the audio data into text, and stores it in a database.

[1662] Step 2:

[1663] Message analysis and emotional state determination

[1664] The server sends the data to the emotion engine for analysis of the received message.

[1665] Input: Communication data in a parsable format

[1666] Data calculation: The emotion engine analyzes the content of the message and determines the user's emotional state (anxiety, fear, etc.).

[1667] Output: Emotional state judgment result

[1668] Specific operation: The server sends a message from the scammer to the emotion engine, which determines that the user is feeling anxious.

[1669] Step 3:

[1670] Generating responses to scammers

[1671] The server sends prompts to the generative AI model, which generates appropriate responses to the imposter.

[1672] Input: Emotional state judgment result, prompt sentence

[1673] Data computation: A generative AI model generates a response based on the prompt.

[1674] Output: The generated response

[1675] Specific operation: The server sends the generative AI model a prompt saying, "Generate an appropriate response to the message from the scammer. Message content: 'Please tell us your bank account information.'" The generative AI model then generates a response saying, "We are unable to provide such information."

[1676] Step 4:

[1677] Sending responses and reporting results

[1678] The server sends the generated response to the impostor.

[1679] Input: The generated response

[1680] Data manipulation: Send the response to the fraudster and log the results.

[1681] Output: Response sent to scammer, report to user

[1682] Specific behavior: The server sends the generated response to the fraudster, and then notifies the user that "an appropriate response has been provided to the fraudster's request."

[1683] Step 5:

[1684] Providing safe information

[1685] The server analyzes the content of the exchange with the scammer and determines whether it is safe.

[1686] Input: Your conversation with the scammer

[1687] Data calculation: Analyzes the content of the exchange and extracts information deemed safe.

[1688] Output: Information deemed safe

[1689] Specific operation: The server analyzes the message from the scammer and, if it determines that the information is safe, notifies the user of that information.

[1690] Step 6:

[1691] Emotion engine interrupts interaction

[1692] If the emotion engine determines that the user's emotions are anxiety or fear, it will discontinue the interaction with the scammer.

[1693] Input: User's emotional state

[1694] Data calculation: The emotion engine analyzes the user's emotional state and determines that they are feeling anxious or scared.

[1695] Output: Instruction to terminate the conversation, notification to the user

[1696] Specific behavior: If the emotion engine determines that the user is feeling scared, the server will disconnect the call and notify the user that the call was from a scammer and that they should not answer calls from this number in the future.

[1697] (Application example 3)

[1698] Next, a description will be given of Application Example 3 of Form Example 3. 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."

[1699] In modern society, sophisticated fraudulent activities via telephone and email are on the rise, increasing the risk that users will be deceived by fraudsters' verbal tactics and documents. In addition, users often feel anxious and scared when interacting with fraudsters, which places a heavy mental burden on them. In response to this, an effective system is needed to automatically detect fraudsters' inappropriate requests and misleading information and protect users.

[1700] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 3 is realized by the following means.

[1701] In this invention, the server includes means for detecting anything unusual in the speech techniques or documents of sophisticated scammers who access the server by phone, email, etc., means for suggesting responses to uncover the scammer's content, means for communicating and responding to the scammer, means for connecting the user only with information that is determined to be safe, means for analyzing the user's emotions and detecting anxiety or fear, means for interrupting communication with the scammer and providing safe information, means for evaluating the safety of information from the scammer, and means for generating appropriate responses to the scammer. This makes it possible to automatically detect attacks from scammers and protect users.

[1702] "Means for detecting unusual content" refers to technology that automatically detects unnatural points or inappropriate information contained in a fraudster's speech or documents.

[1703] "Means for suggesting response content to detect the intent of a fraudster" is a technology for detecting the intent of a fraudster and generating an appropriate response.

[1704] "Means of communicating and responding to scammers" refers to technology that communicates with scammers on behalf of the user and takes appropriate action on their behalf.

[1705] "Means of connecting users with only information that has been determined to be safe" is a technology that evaluates the safety of information from scammers and provides users with only information that has been determined to be safe.

[1706] "Means for analyzing the user's emotions and detecting anxiety or fear" refers to technology that analyzes the user's voice and facial expressions to determine whether the user is feeling anxiety or fear.

[1707] "Means to interrupt communication with scammers and provide safe information" refers to technology that automatically interrupts communication with scammers if the user feels anxious or scared, and provides the user with information that will reassure them.

[1708] The "means for assessing the safety of information from fraudsters" is a technique for assessing whether information provided by fraudsters is safe.

[1709] The "means for generating an appropriate response to an imposter" is a technique for automatically generating an appropriate response to an imposter.

[1710] The system for carrying out the present invention includes a series of means for automatically detecting calls and emails from fraudsters and protecting users. A specific embodiment of this system will be described below.

[1711] Hardware and software used

[1712] Hardware: Smartphones, smart glasses

[1713] Software: emotion engines, generative AI models (e.g., GPT-4), speech recognition systems (e.g., Google Speech-to-Text)

[1714] Data processing and calculation

[1715] 1. Speech Recognition:

[1716] The server converts calls and messages from scammers into text using a voice recognition system.

[1717] Usage example: Converting speech to text using the Google Speech-to-Text API.

[1718] 2. Sentiment analysis:

[1719] The server uses an emotion engine to analyze the user's voice and facial expressions to detect anxiety or fear.

[1720] Use case: Analyzing user sentiment in real time using a sentiment engine.

[1721] 3. Response generation using generative AI models:

[1722] The server uses a generative AI model (e.g., GPT-4) to generate appropriate responses to the imposter.

[1723] Usage example: Prompt GPT-4: "Generate a response that ignores requests from scammers."

[1724] 4. Safety assessment:

[1725] The server determines whether the information from the fraudster is safe and provides only information that is deemed safe to the user.

[1726] Use case: Using generative AI models to assess the safety of information from fraudsters.

[1727] 5. Notification and Action:

[1728] The server will interrupt the conversation with the scammer and notify the user that the call was from a scammer and how to deal with the situation.

[1729] Example use: Use the smartphone notification feature to display a warning message to the user.

[1730] Specific examples

[1731] Example prompts to input to the generative AI model

[1732] "Generate a response that ignores requests from scammers"

[1733] "Evaluate the safety of the information provided by the scammer"

[1734] Example scenario

[1735] 1. Speech Recognition:

[1736] When you receive a call from a scammer, your smartphone converts the voice to text.

[1737] For example: "Hello, your bank account is at risk."

[1738] 2. Sentiment analysis:

[1739] If the user is feeling anxious, the emotion engine will detect it.

[1740] Example: The emotion engine detects that the user's voice is trembling.

[1741] 3. Response generation using generative AI models:

[1742] The server uses the generative AI model to generate appropriate responses to the impostor.

[1743] For example: "That information is incorrect. Please check with your bank directly."

[1744] 4. Safety assessment:

[1745] The server evaluates the security of the information from the fraudster and determines that it is not secure.

[1746] For example, the information "Your bank account is at risk" is considered inappropriate.

[1747] 5. Notification and Action:

[1748] The server will interrupt the conversation with the scammer and send a warning message to the user.

[1749] For example: "This was a scammer calling. Please check with your bank directly."

[1750] In this way, the fraud prevention assistant can automatically detect attacks from fraudsters and protect users.

[1751] The flow of the specific processing in Application Example 3 will be described with reference to FIG.

[1752] Step 1:

[1753] The server receives calls and messages from scammers. It takes voice and text data as input, converts the voice data into text using a speech recognition system (e.g., Google Speech-to-Text), and obtains text data as output.

[1754] Step 2:

[1755] The server inputs the text data into an emotion engine to analyze the user's emotions. It receives the text data and the user's voice and facial expression data as input and detects anxiety or fear. It outputs the emotion analysis results.

[1756] Step 3:

[1757] The server inputs the text data into a generative AI model (e.g., GPT-4) to generate an appropriate response to the fraudster. Using the text data and a prompt (e.g., "Generate a response that ignores the fraudster's request") as input, the generative AI model generates a response. The response text is obtained as output.

[1758] Step 4:

[1759] The server evaluates the safety of the information from the fraudster before providing the generated response text to the user. Using the generated response text and the original text data from the fraudster as input, the server evaluates the safety using a generative AI model. The server obtains the safety evaluation result as output.

[1760] Step 5:

[1761] The server provides the response text to the user only if the information from the fraudster is deemed safe based on the safety evaluation result.The server uses the safety evaluation result and the response text as input and notifies the user of the information that is deemed safe.The server obtains the safe information that is provided to the user as output.

[1762] Step 6:

[1763] The server interrupts communication with the fraudster and provides safety information if the user feels anxious or scared based on the result of the user's emotion analysis.The server uses the emotion analysis result and the original text data from the fraudster as input, interrupts communication with the fraudster, and notifies the user with a warning message.The server obtains the warning message to be provided to the user as output.

[1764] 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.

[1765] 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> ) 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.

[1766] Another example of generative AI is Gemini (internet search engine). <url: https: gemini.google.com ?hl="ja">) are mentioned.

[1767] 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.

[1768] [Third embodiment]

[1769] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[1770] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[1771] 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).

[1772] 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.

[1773] 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.

[1774] 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).

[1775] 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.

[1776] 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.

[1777] 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.

[1778] 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.

[1779] 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.

[1780] Next, the specific processing by the specific processing unit 290 of the data processing device 12 will be described.

[1781] "Example 1"

[1782] One embodiment of the present invention provides a means for detecting unusual content in the speech techniques and documents of sophisticated fraudsters who access a user via telephone, email, etc. This means analyzes the speech techniques and documents of the fraudster and determines whether they contain fraudulent elements. For example, it can detect situations such as when a fraudster makes inappropriate requests to a user or when a fraudster provides information that misleads a user.

[1783] "Example 2"

[1784] Also provided is a means for suggesting responses to uncover the content of a fraudster. This means generates appropriate responses to the fraudster's speech or document content and provides them to the user. For example, if a fraudster makes an inappropriate request to the user, it is possible to suggest a response that ignores the request. Also, if a fraudster provides information that misleads the user, it is possible to suggest a response that denies that information.

[1785] "Example 3"

[1786] Furthermore, a means for communicating and responding to fraudsters is also provided. This means communicates with fraudsters on behalf of the user and reports the results to the user. For example, if a fraudster makes an inappropriate request of the user, the user can respond to the fraudster by ignoring the request. Also, if a fraudster provides information that misleads the user, the user can respond to the fraudster by denying that information.

[1787] "Example 4"

[1788] Finally, a method is provided to connect users only with content that is determined to be safe. This method provides content to users only if the scammer's speech or document content is determined to be safe as a result of interactions with the scammer. For example, if the scammer provides appropriate information to the user, that information can be provided to the user. Also, if the scammer makes inappropriate requests of the user or provides information that is misleading, such information will not be provided to the user.

[1789] The processing flow of each embodiment will be described below.

[1790] "Example 1"

[1791] Step 1: A method for detecting unusual content is activated in the speech techniques and documents of sophisticated fraudsters who access the victim via telephone, email, etc. This method analyzes the speech techniques and contents of documents of fraudsters and determines whether they contain any fraudulent elements.

[1792] Step 2: Generate appropriate responses to the scammer's speech and document content and provide them to the user. In this step, if the scammer makes an inappropriate request to the user, a response that ignores the request is suggested.

[1793] Step 3: Interact with the fraudster on behalf of the user and report the results to the user. In this step, if the fraudster makes an inappropriate request to the user, the system responds to the fraudster by ignoring the request.

[1794] Step 4: If the fraudster's language and document content are deemed safe after interaction with the fraudster, the content is provided to the user. In this step, if the fraudster has provided the user with appropriate information, the information is provided to the user.

[1795] "Example 2"

[1796] Step 1: A method for detecting unusual content is activated in the speech techniques and documents of sophisticated fraudsters who access the victim via telephone, email, etc. This method analyzes the speech techniques and contents of documents of fraudsters and determines whether they contain any fraudulent elements.

[1797] Step 2: Generate appropriate responses to the scammer's speech and document content and provide them to the user. In this step, if the scammer makes an inappropriate request to the user, a response that ignores the request is suggested.

[1798] Step 3: Interact with the fraudster on behalf of the user and report the results to the user. In this step, if the fraudster makes an inappropriate request to the user, the system responds to the fraudster by ignoring the request.

[1799] Step 4: If the fraudster's language and document content are deemed safe after interaction with the fraudster, the content is provided to the user. In this step, if the fraudster has provided the user with appropriate information, the information is provided to the user.

[1800] "Example 3"

[1801] Step 1: A method for detecting unusual content is activated in the speech techniques and documents of sophisticated fraudsters who access the victim via telephone, email, etc. This method analyzes the speech techniques and contents of documents of fraudsters and determines whether they contain any fraudulent elements.

[1802] Step 2: Generate appropriate responses to the scammer's speech and document content and provide them to the user. In this step, if the scammer makes an inappropriate request to the user, a response that ignores the request is suggested.

[1803] Step 3: Interact with the fraudster on behalf of the user and report the results to the user. In this step, if the fraudster makes an inappropriate request to the user, the system responds to the fraudster by ignoring the request.

[1804] Step 4: If the fraudster's language and document content are deemed safe after interaction with the fraudster, the content is provided to the user. In this step, if the fraudster has provided the user with appropriate information, the information is provided to the user.

[1805] "Example 4"

[1806] Step 1: A method for detecting unusual content is activated in the speech techniques and documents of sophisticated fraudsters who access the victim via telephone, email, etc. This method analyzes the speech techniques and contents of documents of fraudsters and determines whether they contain any fraudulent elements.

[1807] Step 2: Generate appropriate responses to the scammer's speech and document content and provide them to the user. In this step, if the scammer makes an inappropriate request to the user, a response that ignores the request is suggested.

[1808] Step 3: Interact with the fraudster on behalf of the user and report the results to the user. In this step, if the fraudster makes an inappropriate request to the user, the system responds to the fraudster by ignoring the request.

[1809] Step 4: If the fraudster's language and document content are deemed safe after interaction with the fraudster, the content is provided to the user. In this step, if the fraudster has provided the user with appropriate information, the information is provided to the user.

[1810] Example 1

[1811] Next, a description will be given of Example 1 of Form 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."

[1812] In modern society, there has been an increase in the number of sophisticated fraud methods carried out via telephone and email. This has increased the risk of many users becoming victims of fraud. Conventional methods have made it difficult to detect and respond to fraud, requiring users to rely on their own judgment to spot fraud. For this reason, there is a need for a system that automates fraud detection and response and protects users.

[1813] 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.

[1814] In this invention, the server includes a means for a user to input a message, a means for the terminal to send the message to the server, a means for the server to analyze the message, a means for the server to determine fraudulent elements using a generative AI model, a means for the server to generate an evaluation result, a means for the server to send the evaluation result to the terminal, and a means for the terminal to display the evaluation result to the user. This automates fraud detection and response, making it possible to reduce the risk of users becoming victims of fraud.

[1815] "Telephone and email" refers to a method of conveying information to users using communication means.

[1816] A "sophisticated fraudster" is someone who uses sophisticated verbal and written techniques to deceive users.

[1817] "Methods for detecting unusual content" are methods for finding unusual or unnatural elements in a message.

[1818] The "means for suggesting response content to detect fraudulent content" is a method for generating appropriate responses to potentially fraudulent messages.

[1819] "Methods of interacting with and responding to scammers" refers to methods for managing communications with scammers and responding appropriately.

[1820] "Means of connecting users only with messages that are deemed safe" is a method of providing users with only messages that are deemed to pose no risk of fraud.

[1821] "Means for users to input messages" refers to the method by which users input messages they receive into the system.

[1822] "Means for a terminal to send a message to a server" refers to a method by which a user's device sends an input message to a server.

[1823] The "means by which the server analyzes the message" refers to a method by which the server analyzes the message received using a natural language processing tool.

[1824] The "means by which the server uses the generative AI model to determine fraudulent elements" is the method by which the server uses the generative AI model to evaluate the fraudulent elements of a message.

[1825] The "means by which the server generates the evaluation result" refers to the method by which the server creates the evaluation result for the message based on the output of the generative AI model.

[1826] The "means by which the server transmits the evaluation results to the terminal" refers to a method by which the server transmits the evaluation results generated to the user's device.

[1827] The "means by which the terminal displays the evaluation results to the user" refers to a method by which the user's device displays the evaluation results it has received on its screen.

[1828] This invention is a system that protects users by detecting any unusual content in the speech techniques and documents of sophisticated fraudsters who access them via telephone, email, etc. This system is realized using a server, terminals, a generative AI model, and a natural language processing tool.

[1829] Hardware and software used

[1830] Hardware: Servers, devices (PCs, smartphones, etc.)

[1831] Software: Generative AI models (e.g., GPT-4), natural language processing tools (e.g., spaCy, NLTK)

[1832] Specific operation of the system

[1833] 1. The user types a message

[1834] Users input the contents of phone calls and emails they receive into the system. For example, if a user receives an email that they suspect is fraudulent, they can copy the contents of the email and paste it into the system's input field.

[1835] 2. The device sends a message to the server

[1836] The device (user's PC or smartphone) sends the entered message to the server. Specifically, it sends the message data to the server using an HTTP request.

[1837] 3. The server parses the message

[1838] The server analyzes the received message using natural language processing tools (e.g., spaCy or NLTK). Specifically, it analyzes the grammatical structure of the message and extracts important keywords and phrases.

[1839] 4. The server uses a generative AI model to determine fraudulent elements

[1840] The server uses a generative AI model (e.g., GPT-4) to determine whether the message content contains fraudulent elements. Specifically, the extracted keywords and phrases are incorporated into a prompt sentence and input into the generative AI model. For example, the prompt sentence "Does this message contain fraudulent elements?" is input into the generative AI model.

[1841] 5. The server generates the evaluation result

[1842] The server evaluates whether the message is suspicious based on the output of the generative AI model. Specifically, it analyzes the output of the generative AI model and generates an evaluation result such as "Possible fraud" or "Not likely fraud."

[1843] 6. The server sends the evaluation results to the device.

[1844] The server sends the generated evaluation results to the terminal. Specifically, it sends the evaluation results to the terminal using an HTTP response.

[1845] 7. The device displays the evaluation results to the user.

[1846] The device then displays the evaluation results to the user. Specifically, it displays a warning message on the user's PC or smartphone screen, such as "This message may be fraudulent."

[1847] Specific examples

[1848] Email example

[1849] Input message: "Congratulations! You are the winner of 1 million yen. Click this link for more details."

[1850] Prompt Sentence Examples

[1851] Prompt: "Does this message contain any elements of deception?"

[1852] Example output

[1853] Generative AI model output: "This message is fraudulent, especially since it asks you to click on a link."

[1854] Evaluation results

[1855] Server rating: "Possibly a scam"

[1856] Display results

[1857] Device displays: "This message may be fraudulent. Do not click on any links."

[1858] In this way, users can detect potentially fraudulent messages in advance and prevent damage before it occurs. This system automates fraud detection and response, reducing the risk of users becoming victims of fraud.

[1859] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1860] Step 1:

[1861] The user enters a message.

[1862] Input: The contents of phone calls and emails received by the user.

[1863] What happens: A user copies and pastes a suspected fraudulent message into an input field on the system.

[1864] Output: The input message data.

[1865] Step 2:

[1866] The device sends a message to the server.

[1867] Input: Message data entered by the user.

[1868] Specific operation: The device (user's PC or smartphone) sends message data to the server using an HTTP request.

[1869] Output: The message data sent to the server.

[1870] Step 3:

[1871] The server parses the message.

[1872] Input: The message data sent to the server.

[1873] What it does: The server uses natural language processing tools (e.g., spaCy or NLTK) to analyze the grammatical structure of the message and extract important keywords and phrases.

[1874] Output: Extracted keywords and phrases.

[1875] Step 4:

[1876] The server uses a generative AI model to determine fraudulent elements.

[1877] Input: Extracted keywords or phrases.

[1878] How it works: The server inputs a prompt sentence like "Does this message contain any elements of fraud?" into a generative AI model (e.g., GPT-4) and obtains the model's output.

[1879] Output: The output of the generative AI model (a rating of whether or not there is fraud).

[1880] Step 5:

[1881] The server generates the evaluation results.

[1882] Input: The output of a generative AI model.

[1883] Specific operation: The server analyzes the output of the generative AI model and generates an evaluation result such as "Possible fraud" or "Not likely to be fraudulent."

[1884] Output: The evaluation result.

[1885] Step 6:

[1886] The server transmits the evaluation results to the terminal.

[1887] Input: Evaluation result.

[1888] Specific operation: The server sends the evaluation results to the terminal using an HTTP response.

[1889] Output: Evaluation results sent to the device.

[1890] Step 7:

[1891] The terminal displays the evaluation results to the user.

[1892] Input: The evaluation result sent to the device.

[1893] Specific behavior: The device displays a warning message on the user's PC or smartphone screen, such as "This message may be fraudulent."

[1894] Output: The evaluation results displayed to the user.

[1895] (Application example 1)

[1896] Next, a description will be given of Application Example 1 of Form 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."

[1897] In recent years, sophisticated frauds using telephone and email have been increasing, increasing the risk of users becoming victims of fraud. In particular, it is difficult for users to detect frauds on their own when fraudsters use sophisticated speaking and writing techniques. Therefore, effective measures are needed to reduce the risk of fraud and ensure the safety of users.

[1898] 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.

[1899] In this invention, the server includes means for detecting anything unusual in the speech techniques or documents of sophisticated fraudsters who access the server by phone or email, means for suggesting responses to uncover the fraudster's content, means for communicating and responding to the fraudster, means for connecting only those determined to be safe to the user, means for analyzing voice data and detecting fraudulent elements, means for analyzing text data and detecting fraudulent elements, and means for issuing a warning if there is a possibility of fraud. This reduces the risk that users will be involved in fraud and ensures safe communication.

[1900] "Means for detecting unusual content" refers to a function that analyzes information received via telephone, email, etc., and automatically detects fraudulent elements or unnatural points.

[1901] The "means for suggesting response content to detect the content of a fraudster" is a function that generates appropriate responses to the fraudster's speech or writing and suggests them to the user.

[1902] "Means of interacting and responding to scammers" refers to functions for managing communication with scammers and taking appropriate action.

[1903] "Means of connecting users with only those things that are determined to be safe" is a function that provides users with only information and communications that are determined to pose no risk of fraud.

[1904] "Means for analyzing voice data and detecting fraudulent elements" refers to a function that analyzes voice data such as telephone conversations and detects elements that may be fraudulent.

[1905] "Means for analyzing text data and detecting fraudulent elements" refers to a function that analyzes text data such as emails and messages and detects elements that may be fraudulent.

[1906] "Means for issuing warnings in the event of possible fraud" is a function that issues a warning to the user when a risk of fraud is detected.

[1907] A system for implementing this invention includes means for detecting anything unusual in the speech techniques or documents of sophisticated fraudsters who access the system by phone or email, means for suggesting responses to uncover the fraudster's content, means for communicating and responding to fraudsters, means for connecting users only with those determined to be safe, means for analyzing voice data and detecting fraudulent elements, means for analyzing text data and detecting fraudulent elements, and means for issuing a warning if there is a possibility of fraud.

[1908] The server uses a speech recognition library to analyze the audio data and converts the audio file into text. The converted text is then analyzed by a generative AI model (e.g., a model using the transformers library) to determine whether it contains fraudulent elements. If fraudulent elements are detected, the server issues a warning to the user.

[1909] In addition, text data (e.g., emails and messages) is also analyzed by the generative AI model to determine whether it contains fraudulent elements. If fraudulent elements are detected, the server issues a warning to the user.

[1910] For example, if a user receives a message saying, "Please tell me your account information," the system will analyze the message and issue a warning if it determines there is a possibility of fraud. An example of a prompt sentence to input into the generative AI model is, "Please tell me your account information."

[1911] This system allows users to reduce the risk of being involved in fraud and ensure safe communication.

[1912] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1913] Step 1:

[1914] The server obtains the voice data or text data received from the user.

[1915] Input: Audio or text data

[1916] Output: Acquired audio data or text data

[1917] Step 2:

[1918] The server uses a speech recognition library to convert the voice data into text.

[1919] Input: Audio data

[1920] Output: Converted text data

[1921] Specific operation: Reads an audio file, performs speech recognition, and outputs it as text.

[1922] Step 3:

[1923] The server inputs the converted text data or directly received text data into the generative AI model.

[1924] Input: Text data

[1925] Output: Analysis results from the generative AI model

[1926] What it does: Text data is fed into a generative AI model (e.g., a model using the transformers library) and analyzed for deceptive elements.

[1927] Step 4:

[1928] The server determines whether the data contains any fraudulent elements based on the analysis results of the generative AI model.

[1929] Input: Analysis results of the generative AI model

[1930] Output: Judgment result on whether there is any fraudulent element

[1931] What it does: Evaluate the analysis results and flag any fraudulent elements detected.

[1932] Step 5:

[1933] The server will warn the user if any fraudulent elements are detected.

[1934] Input: Judgment result regarding the presence or absence of fraudulent elements

[1935] Output: Warning message

[1936] What it does: Sends a warning message to the user if there is a possibility of fraud.

[1937] Step 6:

[1938] If the server does not detect any fraudulent elements, it determines that the transaction is safe and proceeds with normal processing for the user.

[1939] Input: Judgment result regarding the presence or absence of fraudulent elements

[1940] Output: Normal processing result

[1941] What it does: Provide users with information or communications that are determined to be free of fraud risks.

[1942] Example 2

[1943] Next, a description will be given of Example 2 of Form 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."

[1944] In modern society, sophisticated fraud methods via telephone and email are on the rise. This increases the risk that many users will be deceived by the scammers' verbal tactics and documents, and lose their personal information and money. With conventional methods, it is difficult to properly analyze the scammers' messages and respond quickly and accurately, making it difficult to ensure the safety of users.

[1945] 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.

[1946] In this invention, the server includes means for receiving messages from fraudsters, means for transmitting the received messages to the server, means for inputting the messages received by the server into a generative AI model, means for the generative AI model to analyze the fraudster's message and generate an appropriate response content, means for transmitting the generated response content to the user's terminal, means for the user to check the suggested response content and reply to the fraudster as necessary, and means for connecting only those messages that are determined to be safe to the user. This makes it possible to respond quickly and accurately to fraudster's messages and ensure the safety of users.

[1947] A "fraudster" is someone who attempts to fraudulently obtain money or information by deceiving others.

[1948] A "message" is a communication sent by a fraudster to a user, and may be in the form of text, audio, images, etc.

[1949] A "server" is a computer system that processes and stores data over a network.

[1950] A "terminal" is a device that is directly operated by a user, and includes smartphones, tablets, personal computers, etc.

[1951] A "generative AI model" is an algorithm that uses artificial intelligence technology to generate and analyze text, such as a natural language processing model.

[1952] A "prompt" is an instruction entered into a generative AI model, which serves as a trigger for a specific response or analysis.

[1953] The "response content" is the message generated by the generative AI model as a result of analysis, and includes an appropriate response to the scammer.

[1954] "Analysis" is the process by which a generative AI model understands the input message and extracts meaning.

[1955] "Things that have been determined to be safe" refer to communication content and parties that have been determined to pose no risk of fraud.

[1956] "User" means an individual or organization that uses the system to respond to messages from scammers.

[1957] MODE FOR CARRYING OUT THE INVENTION

[1958] The present invention provides a system that generates appropriate responses to messages from fraudsters and helps users respond safely. A specific embodiment of this system is described below.

[1959] Hardware and software used

[1960] Hardware: Servers, terminals

[1961] Software: Generative AI models (e.g., GPT-4)

[1962] System Overview

[1963] The system receives messages from fraudsters, analyzes the messages, and generates appropriate responses. The generated responses are provided to the user, who can then use them to respond to the fraudster.

[1964] Specific actions

[1965] 1. The user receives a message from a scammer

[1966] A user receives a message from a scammer. For example, the scammer asks for your bank details.

[1967] 2. The device sends the received message to the server

[1968] The device sends the received message to the server. Specifically, the device application sends the message to the server via the API.

[1969] 3. The server inputs the received message into the generative AI model.

[1970] The server inputs the received message into the generative AI model. Specifically, the server passes the message to the generative AI model as a prompt. For example, the prompt might be, "I received a message from a scammer saying, 'Please tell me your bank account information.' Please suggest an appropriate response to this message."

[1971] 4. A generative AI model analyzes the scammer's message and generates an appropriate response.

[1972] The generative AI model analyzes the prompt and generates an appropriate response to the scammer's message, for example, "Sorry, but we can't give out any personal information. Let us know if there's anything else we can help you with."

[1973] 5. The server sends the generated response to the user's device.

[1974] The server sends the response received from the generated AI model to the user's device. Specifically, the server sends the response to the device via API.

[1975] 6. The user reviews the suggested responses and responds to the scammer if necessary

[1976] The user checks the suggested responses on their device. The user then uses the suggested responses to reply to the scammer. For example, the user might reply, "Sorry, but I can't give out any personal information."

[1977] Specific examples

[1978] For example, if a user receives the message "Please tell me your bank account information," the following prompt sentence is input into the generative AI model:

[1979] Example prompt sentence:

[1980] I received a message from a scammer asking me to provide my bank details. What would be a good response to this message?

[1981] Example response of the generative AI model:

[1982] "Sorry, I can't give out any personal information. Let me know if there's anything else I can help you with."

[1983] In this way, when a user receives a message from a scammer, the system uses a generative AI model to suggest an appropriate response, helping the user respond safely.

[1984] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1985] Step 1:

[1986] The user receives a message from the scammer.

[1987] Input: A message from the scammer (e.g., "Please tell me your bank details")

[1988] Output: Received message

[1989] What happens: The user's device receives and displays a message from the scammer.

[1990] Step 2:

[1991] The terminal sends the received message to the server.

[1992] Input: Received message

[1993] Output: Message sent to the server

[1994] Specific operation: The device application sends the received message to the server via API.

[1995] Step 3:

[1996] The server inputs the received message into the generative AI model.

[1997] Input: The message sent to the server

[1998] Output: The prompt input to the generative AI model

[1999] Specific operation: The server passes the received message to the generative AI model as a prompt. For example, the prompt might be, "I received a message from a scammer saying, 'Please tell me your bank account information.' Please suggest an appropriate response to this message."

[2000] Step 4:

[2001] A generative AI model analyzes the scammer's message and generates an appropriate response.

[2002] Input: prompt statement

[2003] Output: The generated response

[2004] How it works: The generative AI model analyzes the prompt and generates an appropriate response to the scammer's message. For example, the generative AI model might generate a response like, "Sorry, we can't give out any personal information. Let us know if there's anything else we can help you with."

[2005] Step 5:

[2006] The server sends the generated response to the user's terminal.

[2007] Input: Generated response content

[2008] Output: The response sent to the user's device

[2009] Specific operation: The server sends the response received from the generated AI model to the user's device via API.

[2010] Step 6:

[2011] The user reviews the suggested responses and responds to the scammer if desired.

[2012] Input: The response sent to the user's device

[2013] Output: Reply to the scammer

[2014] What happens: The user reviews the suggested responses on their device and responds to the scammer if necessary. For example, the user might reply, "Sorry, I can't give out my personal information."

[2015] (Application example 2)

[2016] Next, a description will be given of Application Example 2 of Form 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."

[2017] In recent years, sophisticated fraudulent activities using telephone and email have been increasing, increasing the risk of users being deceived by the scammers' verbal techniques and documents. There is a need for a method that allows users to respond appropriately to such fraudulent activities. There is also a need for a system that can detect signs of fraud in real time and provide appropriate responses.

[2018] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for detecting anything unusual in the speech techniques or documents of a clever fraudster who accesses the server by phone, email, etc., means for suggesting responses to uncover the fraudster's content, means for communicating and responding to the fraudster, means for connecting only those determined to be safe to the user, means for recording audio and converting the audio to text, means for determining the possibility of fraud using a generative AI model and generating an appropriate response, and means for displaying the generated response to the user. This allows users to reduce the risk of fraud and communicate safely.

[2019] "Telephone, email, etc." refers to means of communication such as voice calls and email.

[2020] A "sophisticated fraudster" is someone who uses sophisticated verbal and written techniques to deceive users.

[2021] "Means for detecting unusual content" refers to technology that automatically identifies unusual content, including indicators of fraud.

[2022] "Means for suggesting response content to detect fraudsters' content" refers to technology for generating appropriate responses to fraudsters' requests and information and providing them to users.

[2023] "Methods of interacting with and responding to fraudsters" refers to techniques used to manage communications with fraudsters and respond appropriately.

[2024] "Means of connecting users only with communications that are determined to be safe" refers to technology that provides users with only communications that are determined to be void of fraud risks.

[2025] "Means for recording audio and converting audio to text" refers to technology for recording audio data and converting that audio into text information.

[2026] "Means for using generative AI models to determine likelihood of fraud and generate appropriate responses" refers to technologies for using artificial intelligence models to assess likelihood of fraud and generate appropriate responses.

[2027] The term "means for displaying the generated response to the user" refers to a technique for visually presenting the generated response content to the user.

[2028] The system for implementing this invention has the function of detecting anything unusual in the speech techniques or documents of clever fraudsters who access the system by telephone or email, proposing responses to uncover the fraudster's content, communicating and responding to the fraudster, and connecting only those deemed safe to the user.

[2029] System configuration

[2030] The system consists of the following main components:

[2031] 1. Audio Recording Device: Use your smartphone's microphone to record audio.

[2032] 2. Speech Recognition Software: Converts recorded speech into text using a speech recognition library.

[2033] 3. Generative AI model: Uses OpenAI's API to analyze text data and determine the likelihood of fraud.

[2034] 4. Response Generator: Uses generative AI models to generate appropriate responses in cases of potential fraud.

[2035] 5. Display device: The smartphone display is used to display the generated response to the user.

[2036] Processing flow

[2037] 1. Voice recording and transcription: When a user answers a phone call, a voice recorder records the conversation, which is then converted into text by speech recognition software.

[2038] 2. Fraud Detection: A generative AI model analyzes text data to determine the likelihood of fraud, for example, using the following prompt:

[2039] Determine whether the following text is a scam and suggest an appropriate response:

[2040] Please tell us your bank account details

[2041] 3. Generate and display a response: If a fraudulent request is detected, the generative AI model generates an appropriate response and displays it to the user via a display device. For example, a response such as "Please ignore this request. It is extremely dangerous to give your bank account information to others" could be generated.

[2042] Specific examples

[2043] If a user receives the message "Please tell us your bank account information," the system operates as follows:

[2044] 1. A voice recorder records your conversation, which speech recognition software converts into text: "Please tell me your bank account details."

[2045] 2. A generative AI model analyzes this text and determines that it is likely to be fraudulent.

[2046] 3. The response generating device generates a response saying, "Please ignore this request. It is extremely dangerous to give your bank account information to others," and displays it to the user on the display device.

[2047] In this way, users can communicate safely and with less risk of fraud.

[2048] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[2049] Step 1:

[2050] When the user answers the phone, the device's voice recorder records the conversation. The input is the audio of the conversation between the user and the fraudster, and the output is the recorded audio data, which is used in subsequent processing steps.

[2051] Step 2:

[2052] The device's speech recognition software (speech recognition library) converts the recorded voice data into text. The input is the recorded voice data, and the output is text data. This text data is used for fraud detection.

[2053] Step 3:

[2054] The server's generative AI model (OpenAI API) analyzes the text data and determines the likelihood of fraud. The input is the text data, and the output is the evaluation result regarding the likelihood of fraud. Specifically, the analysis is performed using the following prompt sentence:

[2055] Determine whether the following text is a scam and suggest an appropriate response:

[2056] Please tell us your bank account details

[2057] Step 4:

[2058] The server's response generator generates an appropriate response based on the evaluation results of the generative AI model. The input is the evaluation result regarding the possibility of fraud, and the output is the generated response text. For example, a response might be generated that reads, "Please ignore this request. It is very dangerous to give your bank account information to others."

[2059] Step 5:

[2060] The display device of the terminal displays the generated response text to the user. The input is the generated response text, and the output is the response information visually provided to the user. This allows users to communicate safely, reducing the risk of fraud.

[2061] Example 3

[2062] Next, a third embodiment of the third embodiment will be described. 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."

[2063] There is a problem that fraudsters try to deceive users by using clever speech techniques and documents over the phone, email, etc. Conventional systems have difficulty detecting fraudsters' fraudulent activities and responding appropriately, which increases the risk of users becoming victims of fraud. In addition, there is a lack of means to automate interactions with fraudsters and provide users with only safe information. This requires an effective way to prevent fraudsters' fraudulent activities while ensuring the safety of users.

[2064] The specific processing by the specific processing unit 290 of the data processing device 12 in the third embodiment is realized by the following means.

[2065] In this invention, the server includes means for receiving messages from fraudsters, means for analyzing the received messages, means for generating responses to the fraudsters using a generative AI model, means for sending the generated responses to the fraudsters, means for saving the results of interactions with the fraudsters, means for extracting safe information from the results of the interactions, and means for providing the safe information to users. This makes it possible to automatically detect fraudulent activities by fraudsters and respond appropriately. Furthermore, by providing only safe information to users, the safety of users can be ensured and fraud victims can be prevented.

[2066] A "fraudster" is someone who uses fraudulent means to deceive others and obtain money or information illegally.

[2067] A "message" is a communication sent by a scammer to a user, and may be in the form of text, audio, images, etc.

[2068] "Analysis" is the process of analyzing the content of received messages to determine their intent and whether or not they contain fraudulent activity.

[2069] A "generative AI model" is a system that uses artificial intelligence technology to generate and analyze text and speech, such as a natural language processing model.

[2070] A "response" is the reply that the generative AI model creates in response to a message from a scammer.

[2071] "Sending" is the act of sending the generated response to the fraudster over the network.

[2072] "Storage" refers to the act of recording the content of interactions with a fraudster in a storage device such as a database.

[2073] "Extraction" is the process of extracting specific information from the content of stored interactions.

[2074] "Safe information" is information that has been determined to pose no risk to users as a result of interactions with fraudsters.

[2075] "User" means an individual or organization using this system who is to be protected from fraudulent activity by fraudsters.

[2076] The invention is a system that receives messages from fraud...

Claims

1. means for detecting potentially fraudulent information among information communicated to a user; means for suggesting a response to said potentially fraudulent information; means for analyzing the emotional state of the user and providing advice to reduce at least one of the user's stress and anxiety; A system including:

2. The proposed means is provided by a generation system AI; The system of claim 1 .

3. means for interrupting a response to the potentially fraudulent information when the emotion engine determines that the emotion of the user is a negative emotion; The system of claim 1 .

Images