system

Abstract

Provide a system. [Solution] means for receiving a user-entered prompt; a generative artificial intelligence means for generating answers based on the prompts; a morality check artificial intelligence means for evaluating the appropriateness of the generated answers; means for transmitting the answer approved by the morality check artificial intelligence means to a terminal; A system including:

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] Answers generated by generative AI are based on prompts, but due to randomness and prediction errors, they may contain inappropriate answers or erroneous information. Furthermore, malicious users may intentionally input inappropriate prompts and use the resulting answers for fraudulent purposes. This poses a risk of misleading users or spreading inappropriate content. Therefore, a system is needed that can control inappropriate answers and provide users with appropriate and safe information. [Means for solving the problem]

[0005] The present invention aims to solve the above-mentioned problems by providing a system including: means for receiving a prompt input by a user; generative AI means for generating an answer based on the prompt; morality check AI means for evaluating the appropriateness of the generated answer; and means for transmitting the answer approved by the morality check AI means to a terminal. Specifically, the system further includes means for returning a request for correction or regeneration to the generative AI means if the generated answer is determined to be inappropriate, and also includes means for converting the prompt and the generated answer into an appropriate format and means for formatting the answer before displaying it on the terminal, thereby ensuring that the answer provided by the generative AI is appropriate and safe.

[0006] A "prompt" is a question or request that a user enters into a generative artificial intelligence.

[0007] "Generative AI" is an AI system that generates answers based on prompts entered by the user.

[0008] "Moral check artificial intelligence" is an artificial intelligence system that checks the appropriateness of answers generated by generative artificial intelligence and evaluates whether they contain inappropriate content or misinformation.

[0009] A "terminal" is a device, such as a smartphone or computer, through which a user enters prompts and receives generated answers.

[0010] A "means" refers to a method or device used to achieve a particular purpose.

[0011] "Appropriateness" refers to a state in which the generated answer provides appropriate information to the user and does not contain inappropriate expressions or incorrect information.

[0012] "Correction" refers to appropriately changing the content of a generated answer when it is determined to be inappropriate.

[0013] "Regeneration" refers to generating a new answer based on the previous answer when the generated answer is determined to be inappropriate.

[0014] "Formatting" refers to the process of preparing the generated answer in an appropriate display format before displaying it to the user. [Brief explanation of the drawings]

[0015] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram illustrating a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13]FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0016] 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.

[0017] First, the terms used in the following description will be explained.

[0018] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0019] 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.

[0020] 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.

[0021] 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.

[0022] 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."

[0023] [First embodiment]

[0024] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0025] 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.

[0026] 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).

[0027] 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.

[0028] 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.

[0029] 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.

[0030] 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.

[0031] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0032] 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.

[0033] 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.

[0034] 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.

[0035] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0036] The present invention is a system that uses generative artificial intelligence (AI) to generate answers to prompts entered by users and evaluate the appropriateness of those answers, allowing users to receive safe and appropriate information.

[0037] System Configuration

[0038] The system consists of the following main components:

[0039] 1. Input Processing Module

[0040] 2. Generative AI

[0041] 3. Moral Check AI

[0042] 4. Output Processing Module

[0043] User prompt input

[0044] The user uses the terminal to enter a prompt, for example, "Tell me how to adapt to a new environment," and this prompt is sent from the terminal to the server.

[0045] Server-sent prompts

[0046] The device sends the input prompt to the server, which converts it into an appropriate format and passes it to the generative AI.

[0047] Answer generation using generative AI

[0048] The server passes the received prompt to the generative AI, which then generates an answer based on the prompt. For example, the generative AI might generate an answer such as, "In order to adapt to a new environment, it is important to actively communicate with those around you and strengthen self-management."

[0049] Server reception of response

[0050] The server receives the generated answer and then sends the answer to the morale check AI.

[0051] Evaluation by moral check AI

[0052] The server sends the generated answer to the morality check AI, which evaluates the appropriateness of the answer. The morality check AI uses natural language processing technology to check whether the answer contains inappropriate content or misinformation. For example, the morality check AI evaluates the answer and determines that it is appropriate.

[0053] Corrective action (if necessary)

[0054] The server receives the evaluation results from the moral check AI and sends the approved answer directly to the user. On the other hand, if the answer is judged to be inappropriate, it sends a request to the generative AI again based on the moral check AI's suggested corrections, and a new answer is generated.

[0055] Server processing of final response

[0056] The server receives the final approved response and formats it in an appropriate display format in the output processing module.

[0057] User-submitted answers

[0058] The server sends the final answer to the user's device, for example, by sending the answer data in an HTTP response.

[0059] Show Answers

[0060] The device displays the received answer to the user in an easy-to-read format on the device's interface (web browser, application, etc.).

[0061] Specific examples

[0062] For example:

[0063] The user types "Please tell me how to adapt to a new environment" into their device. The device sends this prompt to the server. The server passes the prompt to the generative AI, which generates an answer such as "In order to adapt to a new environment, it is important to actively communicate with those around you and strengthen self-management." The server passes the generated answer to the morality check AI, which evaluates the answer and determines it is appropriate. The server sends the approved answer to the user's device, and the device displays the answer to the user.

[0064] This system allows users to quickly receive appropriate information without being aware of the process by which the generative AI generates it.

[0065] The processing flow will be explained below.

[0066] Step 1:

[0067] The user types a prompt into the terminal, for example, "Tell me how to adapt to a new environment."

[0068] Step 2:

[0069] The terminal sends the entered prompt to the server as an HTTP request.

[0070] Step 3:

[0071] The server converts the received prompt into an appropriate format and sends a request to the generative AI API. For example, it converts it into JSON format and passes it to the generative AI.

[0072] Step 4:

[0073] The generative AI generates answers based on prompts received from the server, such as "To adapt to a new environment, it is important to actively communicate with those around you and strengthen self-management."

[0074] Step 5:

[0075] The server receives the answer generated by the generative AI.

[0076] Step 6:

[0077] The server sends the generated answer to the morality check AI, for example, by sending a request to the morality check AI's API to check whether the generated answer is appropriate.

[0078] Step 7:

[0079] The moral check AI analyzes the answers it receives using natural language processing technology and evaluates the appropriateness of the answers, for example, checking to see if they contain inappropriate language or misinformation.

[0080] Step 8:

[0081] The server receives the evaluation results from the moral check AI. Approved answers proceed to the next step, while answers deemed inappropriate are sent to the generative AI for revision, which then generates a new answer.

[0082] Step 9:

[0083] The server then formats the accepted response in an output processing module into an appropriate display format, for example, HTML or JSON.

[0084] Step 10:

[0085] The server sends the final answer to the user's device as an HTTP response.

[0086] Step 11:

[0087] The device displays the received answer to the user, for example, in a web browser or application in an easy-to-read format.

[0088] Example 1

[0089] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0090] Conventional systems using generative AI lack sufficient means to verify whether the answers to user-input instructions are appropriate, resulting in a high risk of providing users with inappropriate content or erroneous information. Furthermore, if an inappropriate answer is generated, there is no mechanism for automatically correcting it, meaning users end up receiving answers with low reliability. This makes it difficult for users to use the system with confidence.

[0091] 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.

[0092] In this invention, the server includes means for receiving a command input by a user, means for transmitting the command to the server, generative AI means for generating an answer based on the command, ethics check AI means for evaluating the appropriateness of the generated answer, means for transmitting the approved answer to the terminal, and means for the terminal to display the answer to the user, thereby enabling the user to quickly receive reliable and appropriate information.

[0093] "User" refers to a person who utilizes the system to input instructions and receive generated responses.

[0094] "Instruction text" refers to text data such as questions or requests that a user inputs into the system.

[0095] "Server" refers to a central computing unit that receives instructions, processes them, and sends the generated answers to the user's terminal.

[0096] "Terminal" refers to a device used by a user to input instructions and receive and display responses. Examples include a personal computer and a smartphone.

[0097] "Generative artificial intelligence means" refers to an artificial intelligence system or program that generates answers based on input instructions.

[0098] "Ethics check artificial intelligence means" refers to an artificial intelligence system or program that evaluates the appropriateness of generated answers and detects inappropriate content or misinformation.

[0099] "Formatting" refers to the process of converting generated answers into a user-readable form.

[0100] A "prompt" is synonymous with an instruction, and refers to a question or request that a user enters into a system.

[0101] This invention is a system that generates appropriate answers to user inputs, verifies the contents of the answers, and provides them to the user. This system consists of the following main components:

[0102] 1. Input Processing Module

[0103] 2. Generative Artificial Intelligence

[0104] 3. Ethics Checking AI

[0105] 4. Output Processing Module

[0106] First, the user inputs an instruction using the terminal. For example, if the user inputs "Tell me how to adapt to a new environment," this instruction is sent from the terminal to the server.

[0107] Specifically, the device sends the input instruction text to the server as an HTTP POST request. The server then passes the received instruction text to a generative AI (e.g., OpenAI (registered trademark) GPT series). The generative AI generates an answer based on the instruction text. For example, the answer generated might be, "In order to adapt to a new environment, it is important to actively communicate with those around you and strengthen self-management."

[0108] The server then sends the generated answers to an ethics check AI (e.g., a filtering model) that checks whether the generated answers contain inappropriate content or misinformation. This verification reduces the risk of inappropriate answers being provided to users.

[0109] If the generated answer is judged to be inappropriate, the server sends another instruction to the generative AI, requesting it to generate a new answer. This process is repeated until an appropriate answer is generated.

[0110] Finally, the server passes the accepted answer to the output processing module, which formats the answer in a format suitable for the user (for example, converting the generated text into HTML and formatting it so that it is easy for the user to read). Finally, the server sends the formatted answer as an HTTP response to the user's terminal.

[0111] The device then displays the received answers to the user through a web browser or dedicated application interface. This allows the user to quickly receive appropriate and reliable answers without being aware of the generative AI process.

[0112] As a concrete example, a user types "Tell me how to adapt to a new environment" into a terminal, and the terminal sends this to a server. The server passes the prompt to a generative AI, which generates an answer like the one above. The server then passes the generated answer to an ethics check AI for evaluation. If the ethics check AI approves the answer, the server sends it to the terminal, which then displays it to the user.

[0113] This enables systems that make full use of "generative AI models" and "prompt sentences" to provide users with safe and appropriate information.

[0114] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0115] Program processing steps

[0116] Step 1:

[0117] The user uses the terminal to input instructions.

[0118] Input: Text data such as "Tell me how to adapt to a new environment."

[0119] What happens: A user enters text into an input field using a keyboard or voice recognition.

[0120] Step 2:

[0121] The terminal sends the input instruction to the server.

[0122] Input: Instruction text data.

[0123] Output: The HTTP POST request sent to the server.

[0124] Specific behavior: The input text is included in the body of the HTTP request and sent to the specified URL.

[0125] Step 3:

[0126] The server receives the instructions and passes them to the generative artificial intelligence.

[0127] Input: The HTTP POST request sent from the device.

[0128] Output: API request to the generative AI.

[0129] Specific operation: The server extracts instructions from the body of the HTTP request and sends the request to the API of the generative artificial intelligence (e.g., OpenAI GPT series).

[0130] Step 4:

[0131] Generative artificial intelligence generates answers based on instructions.

[0132] Input: Instruction text data.

[0133] Output: The generated answer text.

[0134] How it works: Generative AI uses natural language processing to generate appropriate answers based on the user's instructions. For example, it generates an answer such as, "In order to adapt to a new environment, it is important to actively communicate with those around you and strengthen self-management."

[0135] Step 5:

[0136] The server receives the generated answer and sends it to the ethical check artificial intelligence.

[0137] Input: Answer text received from the generative AI.

[0138] Output: API request to the Ethics Check AI.

[0139] Specific operation: The server extracts the generated answers and sends a request to the API of the ethical check artificial intelligence (e.g., filtering model).

[0140] Step 6:

[0141] Ethics check: Artificial intelligence evaluates the appropriateness of generated answers.

[0142] Input: Answer text data.

[0143] Output: The result of evaluating whether the answer is correct.

[0144] What it does: Ethics check AI analyzes your answers and checks for inappropriate content or misinformation.

[0145] Step 7:

[0146] The server receives the evaluation results of the ethical check artificial intelligence and, if necessary, sends a regeneration request to the generative artificial intelligence.

[0147] Input: Evaluation results from Ethics Check Artificial Intelligence.

[0148] Output: Repeat request to generative AI if necessary.

[0149] Specific operation: If the answer is determined to be inappropriate, the server sends a regeneration request to the generative artificial intelligence to generate a new answer.

[0150] Step 8:

[0151] The server formats the accepted response in the output processing module.

[0152] Input: The accepted answer text.

[0153] Output: The formatted answer text.

[0154] Specific behavior: Converts the accepted answer text into a user-readable format (e.g., HTML format).

[0155] Step 9:

[0156] The server sends the formatted response to the user's terminal.

[0157] Input: The formatted answer text.

[0158] Output: The HTTP response sent to the user's device.

[0159] Specific behavior: Sends formatted text to the terminal as an HTTP response.

[0160] Step 10:

[0161] The terminal displays the received answer to the user.

[0162] Input: The HTTP response received from the server.

[0163] Output: The answer text that is displayed on the screen.

[0164] Specific operation: Display the received data in an easy-to-read format in a web browser or application interface.

[0165] Through these steps, users can receive safe and relevant information quickly.

[0166] (Application example 1)

[0167] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0168] In factories, workers often have questions or problems regarding production lines, machine operation, and quality control, and it is difficult to obtain appropriate answers each time. This reduces work efficiency and increases the risk of operational errors. Furthermore, utilizing human expertise requires a great deal of time and effort.

[0169] 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.

[0170] In this invention, the server includes means for receiving prompts input by a user, generative AI means for generating answers based on the prompts, morality check AI means for evaluating the appropriateness of the generated answers, means for transmitting answers approved by the morality check AI means to an output device, and means for providing answers to questions from factory workers through the output device, thereby making it possible to provide factory workers with prompt and appropriate information in response to their doubts and problems.

[0171] A "user" is a person who uses a system or device and is a worker who is responsible for inputting prompts.

[0172] A "prompt" refers to a question or instruction entered by a user, and is input information used by generative artificial intelligence to generate an answer.

[0173] "Generative AI" is an AI technology that generates appropriate answers based on prompts entered by the user.

[0174] "Moral check AI" is an AI technology that evaluates whether the generated answers are ethically appropriate.

[0175] "Output device" refers to a device for providing the user with the final answer generated by the generative artificial intelligence and the moral check artificial intelligence.

[0176] "Factory worker" means a person who operates production lines and machines, performs quality control, and other tasks within a factory.

[0177] A "question" is a question or inquiry for information that a factory worker has, and is entered as a prompt.

[0178] An "answer" is information generated by the generative AI based on a prompt and provided after being evaluated by the moral check AI.

[0179] This invention is a system that provides prompt and appropriate information to factory workers in response to questions and problems. This system utilizes generative AI and moral check AI to efficiently generate answers and evaluate their appropriateness.

[0180] The system is structured as follows: First, a user inputs a question for a factory worker as a prompt. In response, the input prompt is sent to the server. The server uses generative AI to generate an answer to the prompt. The generated answer is then evaluated for appropriateness by morality check AI.

[0181] The answers that have been approved are sent from the server to an output device and provided to the factory workers. If they are deemed inappropriate, they are regenerated or corrected. For example, OpenAI's API is used for generative AI. Dedicated natural language processing technology is used for moral check AI.

[0182] The specific system operation procedure is as follows: A factory worker speaks to the robot, saying, "Please tell me how to operate this machine." The robot receives this prompt and sends it to the server. The generative AI generates an answer such as, "To operate this machine, first turn it on, then follow the setup guide and press the buttons." The appropriateness of the answer is evaluated by the moral check AI, and if it is deemed appropriate, the robot provides the answer to the factory worker.

[0183] This system allows factory workers to efficiently obtain the appropriate information. Below are some example prompts:

[0184] Example prompt:

[0185] 1. "How do I set up a new production line?"

[0186] 2. "Where are the quality checkpoints for this product?"

[0187] 3. "Please explain how to deal with mechanical problems."

[0188] This will improve work efficiency within the factory and reduce the risk of operational errors and problems.

[0189] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0190] Step 1: The user enters the prompt.

[0191] Input: A factory worker speaks a question to the robot.

[0192] Output: The robot receives the user's prompt and converts it into a data format.

[0193] Specific operation: When a factory worker speaks to the robot, saying, "Please tell me how to operate this machine," the robot's voice recognition function converts the question into text data.

[0194] Step 2: Send the entered prompt from the terminal to the server.

[0195] Input: The prompt converted into text data by the robot.

[0196] Output: Text data is sent to the server.

[0197] Specific operation: The robot converts the prompt text data into an HTTP request format and sends it to the server.

[0198] Step 3: The server generates an answer using generative artificial intelligence.

[0199] Input: The prompt data sent to the server.

[0200] Output: Answer text generated by generative artificial intelligence.

[0201] Specific operation: The server passes the received prompt data to the OpenAI API, and the generative AI model generates an answer such as, "To operate this machine, first turn it on, then press the buttons according to the setup guide."

[0202] Step 4: The generated answers are sent to the moral check AI for evaluation.

[0203] Input: Answer text generated by generative artificial intelligence.

[0204] Output: Evaluation result (approval or disapproval) by the moral check AI.

[0205] Specific operation: The server sends the generated answer to the moral check AI, which uses natural language processing technology to evaluate the appropriateness of the answer and check whether it contains inappropriate content or misinformation.

[0206] Step 5: Revise or regenerate answers based on the results of the moral check AI.

[0207] Input: The generated answer when the evaluation result of the moral check artificial intelligence is disapproval.

[0208] Output: The corrected or regenerated answer text.

[0209] Specific operation: If the moral check AI judges the generated answer to be inappropriate, the server will again prompt the generator AI to generate a new answer. It may also request a regeneration by adding a phrase such as "Please try again."

[0210] Step 6: Moral Check The answer approved by the AI ​​is sent to the terminal.

[0211] Input: Answer text approved by the Morality Check AI.

[0212] Output: The accepted answer is sent to the robot.

[0213] Specific operation: The server sends the final accepted answer to the robot as an HTTP response.

[0214] Step 7: The robot provides the answer to the factory worker.

[0215] Input: The accepted answer text received from the server.

[0216] Output: Provides information to factory workers.

[0217] Specific operation: The robot uses a voice synthesis function to audibly convey the received response to the factory worker. For example, it might say something like, "To operate this machine, first turn it on, then follow the setup guide and press the buttons."

[0218] Through the above steps, factory workers can efficiently obtain appropriate information, thereby improving the accuracy and efficiency of their work.

[0219] 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.

[0220] This invention uses generative artificial intelligence (AI) to generate answers to prompts entered by users, evaluates the appropriateness of those answers, and combines this with an emotion engine that recognizes the user's emotions to provide more accurate and appropriate information. This system allows users to receive safe and appropriate information in a style appropriate to their emotions.

[0221] System Configuration

[0222] The system consists of the following main components:

[0223] 1. Input Processing Module

[0224] 2. Generative AI

[0225] 3. Moral Check AI

[0226] 4. Emotion Engine

[0227] 5. Output Processing Module

[0228] User prompt input

[0229] The user uses the terminal to enter a prompt, for example, "Tell me how to adapt to a new environment," and this prompt is sent from the terminal to the server.

[0230] Server-sent prompts

[0231] The terminal sends the entered prompt to the server as an HTTP request.

[0232] Emotion Analysis

[0233] The server passes the received prompt to the emotion engine, which analyzes the user's emotion. For example, it analyzes that the user is feeling "anxiety" based on the context and keywords of the text.

[0234] Answer generation using generative AI

[0235] The server passes the prompt, including the analyzed emotional information, to the generative AI. The generative AI generates an answer based on the prompt and the emotional information. For example, it generates an answer such as, "To adapt to a new environment, it is important to actively communicate with those around you and strengthen self-management. Don't worry, just take your time to get used to it."

[0236] Server reception of response

[0237] The server receives the generated answer and then sends the answer to the morale check AI.

[0238] Evaluation by moral check AI

[0239] The server sends the generated answer to the morality check AI, which evaluates the appropriateness of the answer. The morality check AI uses natural language processing technology to check whether the answer contains inappropriate content or misinformation. For example, the morality check AI evaluates the answer and determines that it is appropriate.

[0240] Corrective action (if necessary)

[0241] The server receives the evaluation results from the moral check AI and sends the approved answer directly to the user. On the other hand, if the answer is judged to be inappropriate, it sends a request to the generative AI again based on the moral check AI's suggested corrections, and a new answer is generated.

[0242] Server processing of final response

[0243] The server receives the final approved response and formats it in an appropriate display format in the output processing module.

[0244] User-submitted answers

[0245] The server sends the final answer to the user's terminal as an HTTP response.

[0246] Show Answers

[0247] The terminal displays the received answer to the user in an easy-to-read format on the terminal interface (web browser, application, etc.).

[0248] Specific examples

[0249] For example:

[0250] The user types "Tell me how to adapt to a new environment" into their device. The device sends this prompt to the server. The server passes the prompt to the emotion engine, which analyzes the user's emotion as "anxiety." The server passes the prompt along with the analyzed emotion information to the generative AI. The generative AI generates an answer: "To adapt to a new environment, it's important to actively communicate with those around you and strengthen self-management. Don't worry, just get used to it slowly." The server passes the generated answer to the morality check AI, which evaluates the answer and determines it is appropriate. The server sends the approved answer to the user's device, and the device displays the answer to the user.

[0251] This system allows users to quickly and safely receive appropriate information corresponding to their emotions, without being aware of the process that the generative AI is generating.

[0252] The processing flow will be explained below.

[0253] Step 1:

[0254] The user types a prompt into the terminal, for example, "Tell me how to adapt to a new environment."

[0255] Step 2:

[0256] The terminal sends the entered prompt to the server as an HTTP request.

[0257] Step 3:

[0258] The server passes the received prompt to the emotion engine, which analyzes the user's emotion. For example, it analyzes that the user is feeling "anxiety" based on the context and keywords of the text.

[0259] Step 4:

[0260] The server passes the prompt containing the analyzed emotional information to the generative AI. For example, the emotional information "anxiety" is sent along with the prompt.

[0261] Step 5:

[0262] Generative AI generates answers that take emotional information into account, such as, "To adapt to a new environment, it's important to actively communicate with those around you and strengthen self-management. Don't worry, just take your time to get used to it."

[0263] Step 6:

[0264] The server receives the answer generated by the generative AI.

[0265] Step 7:

[0266] The server sends the generated answer to the morality check AI, for example, by sending a request to the morality check AI's API to check whether the generated answer is appropriate.

[0267] Step 8:

[0268] The moral check AI analyzes the answers it receives using natural language processing technology and evaluates the appropriateness of the answers, for example, checking to see if they contain inappropriate language or misinformation.

[0269] Step 9:

[0270] The server receives the evaluation results from the moral check AI. Approved answers proceed to the next step, while answers deemed inappropriate are sent to the generative AI for revision, which then generates a new answer.

[0271] Step 10:

[0272] The server then formats the accepted response in an output processing module into an appropriate display format, for example, HTML or JSON.

[0273] Step 11:

[0274] The server sends the final answer to the user's device as an HTTP response.

[0275] Step 12:

[0276] The device will display the received answer to the user. For example, the answer will be displayed in an easy-to-read format in a web browser or application. Specifically, the device will display a message such as, "To adapt to a new environment, it is important to actively communicate with those around you and strengthen self-management. Don't worry, just take your time to get used to it."

[0277] Example 2

[0278] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0279] Conventional generative AI systems generate answers that do not sufficiently consider the user's emotions, which can result in the system failing to provide the appropriate information the user is looking for. Furthermore, there is a lack of a process for determining the appropriateness of the generated answers, which creates a risk of providing inappropriate information to the user. As a result, there is a risk of the system losing user trust.

[0280] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes means for receiving a prompt input by a user, emotion engine means for analyzing the prompt and recognizing the user's emotion, generative AI means for generating an answer based on the prompt including the analyzed emotion information, morality check AI means for evaluating the appropriateness of the generated answer, means for transmitting the answer approved by the morality check AI means to a terminal, and means for displaying the answer on the terminal. This makes it possible to provide appropriate and safe information according to the user's emotion.

[0281] A "prompt" is text data such as a question or instruction entered by a user.

[0282] An "emotion engine" is software or hardware that analyzes a user's prompts and recognizes the user's emotions from their context and keywords.

[0283] "Generative AI" is an AI model for generating answers in natural language based on prompts and emotional information.

[0284] "Moral check AI" is an AI model that evaluates the appropriateness of answers generated by generative AI and checks whether they contain inappropriate content or misinformation.

[0285] A "terminal" is a device (such as a PC or smartphone) on which a user enters prompts and receives generated answers.

[0286] This invention combines a system that uses generative artificial intelligence (AI) to generate answers to prompts entered by users and evaluates whether the answers are appropriate, with an emotion engine that recognizes the user's emotions. This system allows users to receive safe and appropriate information in a style appropriate to their emotions.

[0287] System Configuration

[0288] The system consists of the following main components:

[0289] 1. Input Processing Module

[0290] 2. Generative AI

[0291] 3. Moral Check AI

[0292] 4. Emotion Engine

[0293] 5. Output Processing Module

[0294] User prompt input

[0295] The user uses the terminal to enter a prompt. Specifically, the user types, "Tell me how to adapt to a new environment." This prompt is sent from the terminal to the server.

[0296] Server-sent prompts

[0297] The terminal sends the entered prompt to the server as an HTTP request.

[0298] Emotion Analysis

[0299] The server passes the received prompt to the emotion engine, which analyzes the user's emotion. The emotion engine, for example, uses a natural language processing toolkit to analyze the context and keywords in the text to determine that the user is feeling "anxiety." This information is converted into JSON format and passed on to the next process.

[0300] Answer generation using generative AI

[0301] The server passes the prompt, including the analyzed emotional information, to the generative AI. Specifically, new JSON data with the emotion engine results added is sent as input to the generative AI. The generative AI then generates an answer based on the prompt and emotional information, using, for example, OpenAI's GPT model. An example of a generated answer would be, "To adapt to a new environment, it's important to actively communicate with those around you and strengthen self-management. Don't worry, just take it easy and get used to it."

[0302] Server reception of response

[0303] The server receives the generated answer and then sends it to the moral check AI. The response from the generating AI comes back in JSON format, which the server parses.

[0304] Evaluation by moral check AI

[0305] The server sends the generated answers to a morality check AI, which evaluates their appropriateness. The morality check AI analyzes the answers created by the generative AI, for example, using a model with specific ethical filters, to check for inappropriate content or misinformation. If the morality check AI determines that the answers are appropriate, this information is also returned in JSON format.

[0306] Corrective action (if necessary)

[0307] The server receives the evaluation results from the morality check AI. If the answer is judged to be inappropriate, it sends a request to the generative AI again based on the correction suggestions provided by the morality check AI to generate a new answer. Specifically, it appropriately combines the original prompt with the morality check AI's feedback and sends a new JSON request to the generative AI.

[0308] Server processing of final response

[0309] The server receives the final approved answer and formats it in an appropriate display format using an output processing module, specifically, formatting the answer in a data format such as HTML or JSON, converting it into a format that is easy for users to understand.

[0310] User-submitted answers

[0311] The server sends the final answer to the user's device as an HTTP response. Specifically, it returns the formatted data to the device again via the HTTPS protocol.

[0312] Show Answers

[0313] The device displays the received response to the user. Specifically, the web browser or application analyzes the received data and displays it in the appropriate location on the screen. The user can see the following text: "To adapt to a new environment, it is important to actively communicate with those around you and strengthen self-management. Don't worry, just take it easy and get used to it."

[0314] This system allows users to quickly and safely receive appropriate information corresponding to their emotions, without being aware of the process that the generative AI is generating.

[0315] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0316] Step 1:

[0317] The user inputs the prompt using their own device. Specifically, the user types "Please tell me how to adapt to the new environment" into the input field of the device's web browser or dedicated application, and clicks the send button. The input data is the prompt text in text format.

[0318] Step 2:

[0319] The terminal sends the entered prompt to the server as an HTTP request. Specifically, JSON format data containing the user input is sent via the HTTPS protocol. The input data is the user prompt, and the output data is the HTTP request sent to the server.

[0320] Step 3:

[0321] The server passes the received prompt to the emotion engine and analyzes the user's emotion. Specifically, the emotion engine (which includes a natural language processing toolkit) analyzes the text and determines from keywords and context that the user is feeling "anxiety." The input data is the received prompt, and the output data is JSON-formatted data containing the analyzed emotion information.

[0322] Step 4:

[0323] The server passes the prompt containing the analyzed emotional information to the generative AI. Specifically, new JSON data with the results of the emotion engine added is sent as input to the generative AI. The generative AI generates an answer based on this. The input data is the prompt containing emotional information, and the output data is JSON format data containing the generated answer.

[0324] Step 5:

[0325] The server receives the generated answer and then sends it to the morality check AI. The response from the generation AI comes back in JSON format, which the server parses. The input data is the generated answer, and the output data is the data sent to the morality check AI.

[0326] Step 6:

[0327] The server sends the generated answer to the morality check AI, which evaluates the appropriateness of the answer. Specifically, the morality check AI (a model with specific ethical filters) analyzes the answer and checks whether it contains inappropriate content or misinformation. If the morality check AI determines that the answer is appropriate, this information is also returned in JSON format. The input data is the generated answer, and the output data is data containing the evaluation result of the appropriateness.

[0328] Step 7:

[0329] The server receives the evaluation results from the morality check AI. If the answer is judged to be inappropriate, it sends a request to the generative AI again based on the correction suggestions provided by the morality check AI to generate a new answer. Specifically, it appropriately combines the original prompt and the morality check AI's feedback and sends a new JSON request to the generative AI. The input data is the evaluation result from the morality check AI, and the output data is the corrected prompt.

[0330] Step 8:

[0331] The server finally receives the approved answer and formats it in the appropriate display format in the output processing module. Specifically, the answer content is formatted in data formats such as HTML or JSON and converted into a form that is easy for users to understand. The input data is the approved answer, and the output data is the formatted answer.

[0332] Step 9:

[0333] The server sends the final answer to the user's terminal as an HTTP response. Specifically, it returns the formatted data to the terminal again via the HTTPS protocol. The input data is the formatted answer, and the output data is the HTTP response.

[0334] Step 10:

[0335] The device displays the received response to the user. Specifically, the web browser or application analyzes the received data and displays it in the appropriate location on the screen. The user can see the following text: "To adapt to a new environment, it is important to actively communicate with those around you and strengthen self-management. Don't worry, just take it easy and get used to it." The input data is the HTTP response, and the output data is the text displayed to the user.

[0336] (Application example 2)

[0337] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0338] Conventional food delivery systems simply present a menu without considering the user's emotions or mood when ordering. This results in a lack of flexible suggestions and advice based on the user's current mental state and emotions, leading to a lack of satisfaction in the user experience. Furthermore, the inability to provide appropriate information based on the user's emotions makes it difficult to provide the comfortable service that users desire.

[0339] 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.

[0340] In this invention, the server includes means for receiving a prompt input by a user, generative AI means for generating an answer based on the prompt, emotion analysis means for providing emotion analysis information to the generative AI means, morality check AI means for evaluating the appropriateness of the generated answer, and means for transmitting to a terminal recommendation information based on the answer approved by the morality check AI means and the emotion analysis information, thereby making it possible to provide the user with appropriate meal menu suggestions and advice in the form of customized messages according to their emotions and moods.

[0341] "User" refers to an individual or organization that uses this system.

[0342] A "prompt" is a request or question that a user enters into a system.

[0343] "Generative artificial intelligence means" refers to AI technologies that generate appropriate answers based on user prompts.

[0344] "Emotion analysis means" refers to a technique for analyzing emotion information from a prompt entered by a user.

[0345] "Moral check artificial intelligence means" refers to AI technology that evaluates the appropriateness of generated answers and eliminates inappropriate content.

[0346] "Terminal" refers to a device for providing generated answers and recommendation information to a user.

[0347] "Emotion analysis information" refers to data that indicates the emotional state of the user extracted by the emotion analysis means.

[0348] "Recommended information" refers to specific options or advice suggested to users based on sentiment analysis information.

[0349] The present invention relates to a system that generates answers to prompts entered by a user, evaluates the appropriateness of the answers, and provides appropriate information based on the user's emotions. In particular, the present invention is applied to food delivery, and provides appropriate menu suggestions and emotional support messages by taking into account the user's emotions when ordering a meal.

[0350] System configuration

[0351] The system consists of the following main components:

[0352] 1. Input Processing Module: A user uses a terminal to input a prompt, for example, the user inputs "I'm feeling down today."

[0353] 2. Generative AI methods: Generate appropriate answers based on user prompts. For example, "If you're looking for something to cheer you up, I recommend chicken soup or salad."

[0354] 3. Emotion analysis: Analyzes emotional information from user input. Example: Analyzes "I feel depressed."

[0355] 4. Moral Check AI Method: Evaluate the appropriateness of the generated answer. For example, check whether the generated answer is harmful to the user’s feelings.

[0356] 5. Output processing module: Sends the final answer and recommendation information to the terminal and displays it. Example: Display "If you're looking for something to cheer you up, we recommend chicken soup or salad."

[0357] Hardware and software used

[0358] Hardware: User devices (smartphones, tablets, etc.), servers

[0359] Software: Python, requests library, emotion analysis module, generative AI module, moral check module, menu recommendation module

[0360] Data processing and calculation

[0361] 1. User prompt input: The terminal receives the user prompt and sends it to the server as an HTTP request.

[0362] 2. Sentiment analysis: The server passes the prompt received to the sentiment analyzer, which analyzes the user's sentiment based on the context and keywords in the text. For example, "I feel depressed" is analyzed and determined to be "negative."

[0363] 3. Answer generation by generative AI: The prompt containing the analyzed emotional information is passed to a generative AI means to generate an answer. The generated answer is in line with the user's emotions. Example: "If you're looking for something to cheer you up, I recommend chicken soup or salad."

[0364] 4. Moral check: The generated answer is sent to the moral check AI means to evaluate its appropriateness. If it is judged to be inappropriate, a correction suggestion is regenerated. Example: The generated answer "Chicken soup or salad is recommended" is judged to be appropriate.

[0365] 5. Sending the final answer: The output processing module formats the accepted answer and sends it to the user's device as an HTTP response. The user's device displays the text "If you're looking for something to cheer you up, I recommend chicken soup or salad."

[0366] Specific examples

[0367] A user uses a food delivery assistant application and inputs, "I'm feeling down today." The device sends this to the server. The server receives the prompt, and the emotion analysis means analyzes the emotional information (negative). Based on the analyzed emotional information, the generative AI generates an answer: "If you're looking for something to cheer you up, I recommend chicken soup or salad." After the moral check AI determines this answer is appropriate, the final answer is sent to the user's device and displayed.

[0368] Example of a user prompt:

[0369] "I'm feeling down today, so I'd like you to recommend a dish that will cheer me up."

[0370] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0371] Step 1:

[0372] The user inputs a prompt through the terminal. Specifically, the user inputs "I'm feeling depressed today." This input text is acquired as prompt information.

[0373] Input: User prompt (e.g., "I'm feeling down today")

[0374] Output: Prompt information (text)

[0375] Step 2:

[0376] The terminal sends the input prompt information to the server as an HTTP request. Specifically, the prompt information is sent to the server's input processing module using a REST API or similar.

[0377] Input: Prompt information (text)

[0378] Output: Prompt information sent to the server (HTTP request)

[0379] Step 3:

[0380] The server passes the received prompt information to the emotion analysis means, which analyzes the user's emotion. Specifically, it uses a text analysis algorithm to recognize the user's emotion (e.g., "negative") from the prompt content.

[0381] Input: Prompt information (text)

[0382] Output: Sentiment analysis information (e.g., "negative")

[0383] Step 4:

[0384] The server passes prompt information, including the analyzed emotional information, to the generative AI means. The generative AI model generates an appropriate response based on the prompt information and emotional information. Specifically, in response to the statement "I'm feeling down today," the AI ​​model generates the response "If you're looking for something to cheer you up, I recommend chicken soup or a salad."

[0385] Input: prompt information (text), sentiment analysis information (e.g., "negative")

[0386] Output: Generated answer (e.g., "For comfort food, chicken soup or salad is recommended.")

[0387] Step 5:

[0388] The server sends the generated answer to the morality check AI means, which evaluates the appropriateness of the answer. Specifically, the morality check AI checks whether the generated answer contains inappropriate content or misinformation, and returns an evaluation result of whether the answer is appropriate.

[0389] Input: Generated answer (text)

[0390] Output: Evaluation result (appropriate / inappropriate)

[0391] Step 6:

[0392] The server receives the evaluation results from the moral check AI, and if the evaluation results are appropriate, it sends the final answer and recommendation information to the device. Specifically, it sends the answer that it judges to be appropriate (e.g., "If you're looking for something to cheer you up, we recommend chicken soup or salad.") to the user's device as an HTTP response.

[0393] Input: Evaluation results, generated answers (text)

[0394] Output: Final answer (text) sent to the terminal

[0395] Step 7:

[0396] The device displays the final answer it received to the user. Specifically, it displays "If you're looking for something to cheer you up, we recommend chicken soup or salad" in an easy-to-read format on the device's UI (user interface).

[0397] Input: Final answer (text)

[0398] Output: what is displayed to the user (text)

[0399] 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.

[0400] 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> ), Gemini (registered trademark) (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0401] 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.

[0402] [Second embodiment]

[0403] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0404] 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.

[0405] 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).

[0406] 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.

[0407] 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.

[0408] 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).

[0409] 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.

[0410] 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.

[0411] 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.

[0412] 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.

[0413] 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.

[0414] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0415] The present invention is a system that uses generative artificial intelligence (AI) to generate answers to prompts entered by users and evaluate the appropriateness of those answers, allowing users to receive safe and appropriate information.

[0416] System Configuration

[0417] The system consists of the following main components:

[0418] 1. Input Processing Module

[0419] 2. Generative AI

[0420] 3. Moral Check AI

[0421] 4. Output Processing Module

[0422] User prompt input

[0423] The user uses the terminal to enter a prompt, for example, "Tell me how to adapt to a new environment," and this prompt is sent from the terminal to the server.

[0424] Server-sent prompts

[0425] The device sends the input prompt to the server, which converts it into an appropriate format and passes it to the generative AI.

[0426] Answer generation using generative AI

[0427] The server passes the received prompt to the generative AI, which then generates an answer based on the prompt. For example, the generative AI might generate an answer such as, "In order to adapt to a new environment, it is important to actively communicate with those around you and strengthen self-management."

[0428] Server reception of response

[0429] The server receives the generated answer and then sends the answer to the morale check AI.

[0430] Evaluation by moral check AI

[0431] The server sends the generated answer to the morality check AI, which evaluates the appropriateness of the answer. The morality check AI uses natural language processing technology to check whether the answer contains inappropriate content or misinformation. For example, the morality check AI evaluates the answer and determines that it is appropriate.

[0432] Corrective action (if necessary)

[0433] The server receives the evaluation results from the moral check AI and sends the approved answer directly to the user. On the other hand, if the answer is judged to be inappropriate, it sends a request to the generative AI again based on the moral check AI's suggested corrections, and a new answer is generated.

[0434] Server processing of final response

[0435] The server receives the final approved response and formats it in an appropriate display format in the output processing module.

[0436] User-submitted answers

[0437] The server sends the final answer to the user's device, for example, by sending the answer data in an HTTP response.

[0438] Show Answers

[0439] The device displays the received answer to the user in an easy-to-read format on the device's interface (web browser, application, etc.).

[0440] Specific examples

[0441] For example:

[0442] The user types "Please tell me how to adapt to a new environment" into their device. The device sends this prompt to the server. The server passes the prompt to the generative AI, which generates an answer such as "In order to adapt to a new environment, it is important to actively communicate with those around you and strengthen self-management." The server passes the generated answer to the morality check AI, which evaluates the answer and determines it is appropriate. The server sends the approved answer to the user's device, and the device displays the answer to the user.

[0443] This system allows users to quickly receive appropriate information without being aware of the process by which the generative AI generates it.

[0444] The processing flow will be explained below.

[0445] Step 1:

[0446] The user types a prompt into the terminal, for example, "Tell me how to adapt to a new environment."

[0447] Step 2:

[0448] The terminal sends the entered prompt to the server as an HTTP request.

[0449] Step 3:

[0450] The server converts the received prompt into an appropriate format and sends a request to the generative AI API. For example, it converts it into JSON format and passes it to the generative AI.

[0451] Step 4:

[0452] The generative AI generates answers based on prompts received from the server, such as "To adapt to a new environment, it is important to actively communicate with those around you and strengthen self-management."

[0453] Step 5:

[0454] The server receives the answer generated by the generative AI.

[0455] Step 6:

[0456] The server sends the generated answer to the morality check AI, for example, by sending a request to the morality check AI's API to check whether the generated answer is appropriate.

[0457] Step 7:

[0458] The moral check AI analyzes the answers it receives using natural language processing technology and evaluates the appropriateness of the answers, for example, checking to see if they contain inappropriate language or misinformation.

[0459] Step 8:

[0460] The server receives the evaluation results from the moral check AI. Approved answers proceed to the next step, while answers deemed inappropriate are sent to the generative AI for revision, which then generates a new answer.

[0461] Step 9:

[0462] The server then formats the accepted response in an output processing module into an appropriate display format, for example, HTML or JSON.

[0463] Step 10:

[0464] The server sends the final answer to the user's device as an HTTP response.

[0465] Step 11:

[0466] The device displays the received answer to the user, for example, in a web browser or application in an easy-to-read format.

[0467] Example 1

[0468] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0469] Conventional systems using generative AI lack sufficient means to verify whether the answers to user-input instructions are appropriate, resulting in a high risk of providing users with inappropriate content or erroneous information. Furthermore, if an inappropriate answer is generated, there is no mechanism for automatically correcting it, meaning users end up receiving answers with low reliability. This makes it difficult for users to use the system with confidence.

[0470] 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.

[0471] In this invention, the server includes means for receiving a command input by a user, means for transmitting the command to the server, generative AI means for generating an answer based on the command, ethics check AI means for evaluating the appropriateness of the generated answer, means for transmitting the approved answer to the terminal, and means for the terminal to display the answer to the user, thereby enabling the user to quickly receive reliable and appropriate information.

[0472] "User" refers to a person who utilizes the system to input instructions and receive generated responses.

[0473] "Instruction text" refers to text data such as questions or requests that a user inputs into the system.

[0474] "Server" refers to a central computing unit that receives instructions, processes them, and sends the generated answers to the user's terminal.

[0475] "Terminal" refers to a device used by a user to input instructions and receive and display responses. Examples include a personal computer and a smartphone.

[0476] "Generative artificial intelligence means" refers to an artificial intelligence system or program that generates answers based on input instructions.

[0477] "Ethics check artificial intelligence means" refers to an artificial intelligence system or program that evaluates the appropriateness of generated answers and detects inappropriate content or misinformation.

[0478] "Formatting" refers to the process of converting generated answers into a user-readable form.

[0479] A "prompt" is synonymous with an instruction, and refers to a question or request that a user enters into a system.

[0480] This invention is a system that generates appropriate answers to user inputs, verifies the contents of the answers, and provides them to the user. This system consists of the following main components:

[0481] 1. Input Processing Module

[0482] 2. Generative Artificial Intelligence

[0483] 3. Ethics Checking AI

[0484] 4. Output Processing Module

[0485] First, the user inputs an instruction using the terminal. For example, if the user inputs "Tell me how to adapt to a new environment," this instruction is sent from the terminal to the server.

[0486] Specifically, the device sends the input instruction text to the server as an HTTP POST request. The server then passes the received instruction text to a generative AI (e.g., the OpenAI GPT series). The generative AI generates an answer based on the instruction text. For example, it might generate an answer such as, "In order to adapt to a new environment, it is important to actively communicate with those around you and strengthen self-management."

[0487] The server then sends the generated answers to an ethics check AI (e.g., a filtering model) that checks whether the generated answers contain inappropriate content or misinformation. This verification reduces the risk of inappropriate answers being provided to users.

[0488] If the generated answer is judged to be inappropriate, the server sends another instruction to the generative AI, requesting it to generate a new answer. This process is repeated until an appropriate answer is generated.

[0489] Finally, the server passes the accepted answer to the output processing module, which formats the answer in a format suitable for the user (for example, converting the generated text into HTML and formatting it so that it is easy for the user to read). Finally, the server sends the formatted answer as an HTTP response to the user's terminal.

[0490] The device then displays the received answers to the user through a web browser or dedicated application interface. This allows the user to quickly receive appropriate and reliable answers without being aware of the generative AI process.

[0491] As a concrete example, a user types "Tell me how to adapt to a new environment" into a terminal, and the terminal sends this to a server. The server passes the prompt to a generative AI, which generates an answer like the one above. The server then passes the generated answer to an ethics check AI for evaluation. If the ethics check AI approves the answer, the server sends it to the terminal, which then displays it to the user.

[0492] This enables systems that make full use of "generative AI models" and "prompt sentences" to provide users with safe and appropriate information.

[0493] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0494] Program processing steps

[0495] Step 1:

[0496] The user uses the terminal to input instructions.

[0497] Input: Text data such as "Tell me how to adapt to a new environment."

[0498] What happens: A user enters text into an input field using a keyboard or voice recognition.

[0499] Step 2:

[0500] The terminal sends the input instruction to the server.

[0501] Input: Instruction text data.

[0502] Output: The HTTP POST request sent to the server.

[0503] Specific behavior: The input text is included in the body of the HTTP request and sent to the specified URL.

[0504] Step 3:

[0505] The server receives the instructions and passes them to the generative artificial intelligence.

[0506] Input: The HTTP POST request sent from the device.

[0507] Output: API request to the generative AI.

[0508] Specific operation: The server extracts instructions from the body of the HTTP request and sends the request to the API of the generative artificial intelligence (e.g., OpenAI GPT series).

[0509] Step 4:

[0510] Generative artificial intelligence generates answers based on instructions.

[0511] Input: Instruction text data.

[0512] Output: The generated answer text.

[0513] How it works: Generative AI uses natural language processing to generate appropriate answers based on the user's instructions. For example, it generates an answer such as, "In order to adapt to a new environment, it is important to actively communicate with those around you and strengthen self-management."

[0514] Step 5:

[0515] The server receives the generated answer and sends it to the ethical check artificial intelligence.

[0516] Input: Answer text received from the generative AI.

[0517] Output: API request to the Ethics Check AI.

[0518] Specific operation: The server extracts the generated answers and sends a request to the API of the ethical check artificial intelligence (e.g., filtering model).

[0519] Step 6:

[0520] Ethics check: Artificial intelligence evaluates the appropriateness of generated answers.

[0521] Input: Answer text data.

[0522] Output: The result of evaluating whether the answer is correct.

[0523] What it does: Ethics check AI analyzes your answers and checks for inappropriate content or misinformation.

[0524] Step 7:

[0525] The server receives the evaluation results of the ethical check artificial intelligence and, if necessary, sends a regeneration request to the generative artificial intelligence.

[0526] Input: Evaluation results from Ethics Check Artificial Intelligence.

[0527] Output: Repeat request to generative AI if necessary.

[0528] Specific operation: If the answer is determined to be inappropriate, the server sends a regeneration request to the generative artificial intelligence to generate a new answer.

[0529] Step 8:

[0530] The server formats the accepted response in the output processing module.

[0531] Input: The accepted answer text.

[0532] Output: The formatted answer text.

[0533] Specific behavior: Converts the accepted answer text into a user-readable format (e.g., HTML format).

[0534] Step 9:

[0535] The server sends the formatted response to the user's terminal.

[0536] Input: The formatted answer text.

[0537] Output: The HTTP response sent to the user's device.

[0538] Specific behavior: Sends formatted text to the terminal as an HTTP response.

[0539] Step 10:

[0540] The terminal displays the received answer to the user.

[0541] Input: The HTTP response received from the server.

[0542] Output: The answer text that is displayed on the screen.

[0543] Specific operation: Display the received data in an easy-to-read format in a web browser or application interface.

[0544] Through these steps, users can receive safe and relevant information quickly.

[0545] (Application example 1)

[0546] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0547] In factories, workers often have questions or problems regarding production lines, machine operation, and quality control, and it is difficult to obtain appropriate answers each time. This reduces work efficiency and increases the risk of operational errors. Furthermore, utilizing human expertise requires a great deal of time and effort.

[0548] 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.

[0549] In this invention, the server includes means for receiving prompts input by a user, generative AI means for generating answers based on the prompts, morality check AI means for evaluating the appropriateness of the generated answers, means for transmitting answers approved by the morality check AI means to an output device, and means for providing answers to questions from factory workers through the output device, thereby making it possible to provide factory workers with prompt and appropriate information in response to their doubts and problems.

[0550] A "user" is a person who uses a system or device and is a worker who is responsible for inputting prompts.

[0551] A "prompt" refers to a question or instruction entered by a user, and is input information used by generative artificial intelligence to generate an answer.

[0552] "Generative AI" is an AI technology that generates appropriate answers based on prompts entered by the user.

[0553] "Moral check AI" is an AI technology that evaluates whether the generated answers are ethically appropriate.

[0554] "Output device" refers to a device for providing the user with the final answer generated by the generative artificial intelligence and the moral check artificial intelligence.

[0555] "Factory worker" means a person who operates production lines and machines, performs quality control, and other tasks within a factory.

[0556] A "question" is a question or inquiry for information that a factory worker has, and is entered as a prompt.

[0557] An "answer" is information generated by the generative AI based on a prompt and provided after being evaluated by the moral check AI.

[0558] This invention is a system that provides prompt and appropriate information to factory workers in response to questions and problems. This system utilizes generative AI and moral check AI to efficiently generate answers and evaluate their appropriateness.

[0559] The system is structured as follows: First, a user inputs a question for a factory worker as a prompt. In response, the input prompt is sent to the server. The server uses generative AI to generate an answer to the prompt. The generated answer is then evaluated for appropriateness by morality check AI.

[0560] The answers that have been approved are sent from the server to an output device and provided to the factory workers. If they are deemed inappropriate, they are regenerated or corrected. For example, OpenAI's API is used for generative AI. Dedicated natural language processing technology is used for moral check AI.

[0561] The specific system operation procedure is as follows: A factory worker speaks to the robot, saying, "Please tell me how to operate this machine." The robot receives this prompt and sends it to the server. The generative AI generates an answer such as, "To operate this machine, first turn it on, then follow the setup guide and press the buttons." The appropriateness of the answer is evaluated by the moral check AI, and if it is deemed appropriate, the robot provides the answer to the factory worker.

[0562] This system allows factory workers to efficiently obtain the appropriate information. Below are some example prompts:

[0563] Example prompt:

[0564] 1. "How do I set up a new production line?"

[0565] 2. "Where are the quality checkpoints for this product?"

[0566] 3. "Please explain how to deal with mechanical problems."

[0567] This will improve work efficiency within the factory and reduce the risk of operational errors and problems.

[0568] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0569] Step 1: The user enters the prompt.

[0570] Input: A factory worker speaks a question to the robot.

[0571] Output: The robot receives the user's prompt and converts it into a data format.

[0572] Specific operation: When a factory worker speaks to the robot, saying, "Please tell me how to operate this machine," the robot's voice recognition function converts the question into text data.

[0573] Step 2: Send the entered prompt from the terminal to the server.

[0574] Input: The prompt converted into text data by the robot.

[0575] Output: Text data is sent to the server.

[0576] Specific operation: The robot converts the prompt text data into an HTTP request format and sends it to the server.

[0577] Step 3: The server generates an answer using generative artificial intelligence.

[0578] Input: The prompt data sent to the server.

[0579] Output: Answer text generated by generative artificial intelligence.

[0580] Specific operation: The server passes the received prompt data to the OpenAI API, and the generative AI model generates an answer such as, "To operate this machine, first turn it on, then press the buttons according to the setup guide."

[0581] Step 4: The generated answers are sent to the moral check AI for evaluation.

[0582] Input: Answer text generated by generative artificial intelligence.

[0583] Output: Evaluation result (approval or disapproval) by the moral check AI.

[0584] Specific operation: The server sends the generated answer to the moral check AI, which uses natural language processing technology to evaluate the appropriateness of the answer and check whether it contains inappropriate content or misinformation.

[0585] Step 5: Revise or regenerate answers based on the results of the moral check AI.

[0586] Input: The generated answer when the evaluation result of the moral check artificial intelligence is disapproval.

[0587] Output: The corrected or regenerated answer text.

[0588] Specific operation: If the moral check AI judges the generated answer to be inappropriate, the server will again prompt the generator AI to generate a new answer. It may also request a regeneration by adding a phrase such as "Please try again."

[0589] Step 6: Moral Check The answer approved by the AI ​​is sent to the terminal.

[0590] Input: Answer text approved by the Morality Check AI.

[0591] Output: The accepted answer is sent to the robot.

[0592] Specific operation: The server sends the final accepted answer to the robot as an HTTP response.

[0593] Step 7: The robot provides the answer to the factory worker.

[0594] Input: The accepted answer text received from the server.

[0595] Output: Provides information to factory workers.

[0596] Specific operation: The robot uses a voice synthesis function to audibly convey the received response to the factory worker. For example, it might say something like, "To operate this machine, first turn it on, then follow the setup guide and press the buttons."

[0597] Through the above steps, factory workers can efficiently obtain appropriate information, thereby improving the accuracy and efficiency of their work.

[0598] 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.

[0599] This invention uses generative artificial intelligence (AI) to generate answers to prompts entered by users, evaluates the appropriateness of those answers, and combines this with an emotion engine that recognizes the user's emotions to provide more accurate and appropriate information. This system allows users to receive safe and appropriate information in a style appropriate to their emotions.

[0600] System Configuration

[0601] The system consists of the following main components:

[0602] 1. Input Processing Module

[0603] 2. Generative AI

[0604] 3. Moral Check AI

[0605] 4. Emotion Engine

[0606] 5. Output Processing Module

[0607] User prompt input

[0608] The user uses the terminal to enter a prompt, for example, "Tell me how to adapt to a new environment," and this prompt is sent from the terminal to the server.

[0609] Server-sent prompts

[0610] The terminal sends the entered prompt to the server as an HTTP request.

[0611] Emotion Analysis

[0612] The server passes the received prompt to the emotion engine, which analyzes the user's emotion. For example, it analyzes that the user is feeling "anxiety" based on the context and keywords of the text.

[0613] Answer generation using generative AI

[0614] The server passes the prompt, including the analyzed emotional information, to the generative AI. The generative AI generates an answer based on the prompt and the emotional information. For example, it generates an answer such as, "To adapt to a new environment, it is important to actively communicate with those around you and strengthen self-management. Don't worry, just take your time to get used to it."

[0615] Server reception of response

[0616] The server receives the generated answer and then sends the answer to the morale check AI.

[0617] Evaluation by moral check AI

[0618] The server sends the generated answer to the morality check AI, which evaluates the appropriateness of the answer. The morality check AI uses natural language processing technology to check whether the answer contains inappropriate content or misinformation. For example, the morality check AI evaluates the answer and determines that it is appropriate.

[0619] Corrective action (if necessary)

[0620] The server receives the evaluation results from the moral check AI and sends the approved answer directly to the user. On the other hand, if the answer is judged to be inappropriate, it sends a request to the generative AI again based on the moral check AI's suggested corrections, and a new answer is generated.

[0621] Server processing of final response

[0622] The server receives the final approved response and formats it in an appropriate display format in the output processing module.

[0623] User-submitted answers

[0624] The server sends the final answer to the user's terminal as an HTTP response.

[0625] Show Answers

[0626] The terminal displays the received answer to the user in an easy-to-read format on the terminal interface (web browser, application, etc.).

[0627] Specific examples

[0628] For example:

[0629] The user types "Tell me how to adapt to a new environment" into their device. The device sends this prompt to the server. The server passes the prompt to the emotion engine, which analyzes the user's emotion as "anxiety." The server passes the prompt along with the analyzed emotion information to the generative AI. The generative AI generates an answer: "To adapt to a new environment, it's important to actively communicate with those around you and strengthen self-management. Don't worry, just get used to it slowly." The server passes the generated answer to the morality check AI, which evaluates the answer and determines it is appropriate. The server sends the approved answer to the user's device, and the device displays the answer to the user.

[0630] This system allows users to quickly and safely receive appropriate information corresponding to their emotions, without being aware of the process that the generative AI is generating.

[0631] The processing flow will be explained below.

[0632] Step 1:

[0633] The user types a prompt into the terminal, for example, "Tell me how to adapt to a new environment."

[0634] Step 2:

[0635] The terminal sends the entered prompt to the server as an HTTP request.

[0636] Step 3:

[0637] The server passes the received prompt to the emotion engine, which analyzes the user's emotion. For example, it analyzes that the user is feeling "anxiety" based on the context and keywords of the text.

[0638] Step 4:

[0639] The server passes the prompt containing the analyzed emotional information to the generative AI. For example, the emotional information "anxiety" is sent along with the prompt.

[0640] Step 5:

[0641] Generative AI generates answers that take emotional information into account, such as, "To adapt to a new environment, it's important to actively communicate with those around you and strengthen self-management. Don't worry, just take your time to get used to it."

[0642] Step 6:

[0643] The server receives the answer generated by the generative AI.

[0644] Step 7:

[0645] The server sends the generated answer to the morality check AI, for example, by sending a request to the morality check AI's API to check whether the generated answer is appropriate.

[0646] Step 8:

[0647] The moral check AI analyzes the answers it receives using natural language processing technology and evaluates the appropriateness of the answers, for example, checking to see if they contain inappropriate language or misinformation.

[0648] Step 9:

[0649] The server receives the evaluation results from the moral check AI. Approved answers proceed to the next step, while answers deemed inappropriate are sent to the generative AI for revision, which then generates a new answer.

[0650] Step 10:

[0651] The server then formats the accepted response in an output processing module into an appropriate display format, for example, HTML or JSON.

[0652] Step 11:

[0653] The server sends the final answer to the user's device as an HTTP response.

[0654] Step 12:

[0655] The device will display the received answer to the user. For example, the answer will be displayed in an easy-to-read format in a web browser or application. Specifically, the device will display a message such as, "To adapt to a new environment, it is important to actively communicate with those around you and strengthen self-management. Don't worry, just take your time to get used to it."

[0656] Example 2

[0657] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0658] Conventional generative AI systems generate answers that do not sufficiently consider the user's emotions, which can result in the system failing to provide the appropriate information the user is looking for. Furthermore, there is a lack of a process for determining the appropriateness of the generated answers, which creates a risk of providing inappropriate information to the user. As a result, there is a risk of the system losing user trust.

[0659] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes means for receiving a prompt input by a user, emotion engine means for analyzing the prompt and recognizing the user's emotion, generative AI means for generating an answer based on the prompt including the analyzed emotion information, morality check AI means for evaluating the appropriateness of the generated answer, means for transmitting the answer approved by the morality check AI means to a terminal, and means for displaying the answer on the terminal. This makes it possible to provide appropriate and safe information according to the user's emotion.

[0660] A "prompt" is text data such as a question or instruction entered by a user.

[0661] An "emotion engine" is software or hardware that analyzes a user's prompts and recognizes the user's emotions from their context and keywords.

[0662] "Generative AI" is an AI model for generating answers in natural language based on prompts and emotional information.

[0663] "Moral check AI" is an AI model that evaluates the appropriateness of answers generated by generative AI and checks whether they contain inappropriate content or misinformation.

[0664] A "terminal" is a device (such as a PC or smartphone) on which a user enters prompts and receives generated answers.

[0665] This invention combines a system that uses generative artificial intelligence (AI) to generate answers to prompts entered by users and evaluates whether the answers are appropriate, with an emotion engine that recognizes the user's emotions. This system allows users to receive safe and appropriate information in a style appropriate to their emotions.

[0666] System Configuration

[0667] The system consists of the following main components:

[0668] 1. Input Processing Module

[0669] 2. Generative AI

[0670] 3. Moral Check AI

[0671] 4. Emotion Engine

[0672] 5. Output Processing Module

[0673] User prompt input

[0674] The user uses the terminal to enter a prompt. Specifically, the user types, "Tell me how to adapt to a new environment." This prompt is sent from the terminal to the server.

[0675] Server-sent prompts

[0676] The terminal sends the entered prompt to the server as an HTTP request.

[0677] Emotion Analysis

[0678] The server passes the received prompt to the emotion engine, which analyzes the user's emotion. The emotion engine, for example, uses a natural language processing toolkit to analyze the context and keywords in the text to determine that the user is feeling "anxiety." This information is converted into JSON format and passed on to the next process.

[0679] Answer generation using generative AI

[0680] The server passes the prompt, including the analyzed emotional information, to the generative AI. Specifically, new JSON data with the emotion engine results added is sent as input to the generative AI. The generative AI then generates an answer based on the prompt and emotional information, using, for example, OpenAI's GPT model. An example of a generated answer would be, "To adapt to a new environment, it's important to actively communicate with those around you and strengthen self-management. Don't worry, just take it easy and get used to it."

[0681] Server reception of response

[0682] The server receives the generated answer and then sends it to the moral check AI. The response from the generating AI comes back in JSON format, which the server parses.

[0683] Evaluation by moral check AI

[0684] The server sends the generated answers to a morality check AI, which evaluates their appropriateness. The morality check AI analyzes the answers created by the generative AI, for example, using a model with specific ethical filters, to check for inappropriate content or misinformation. If the morality check AI determines that the answers are appropriate, this information is also returned in JSON format.

[0685] Corrective action (if necessary)

[0686] The server receives the evaluation results from the morality check AI. If the answer is judged to be inappropriate, it sends a request to the generative AI again based on the correction suggestions provided by the morality check AI to generate a new answer. Specifically, it appropriately combines the original prompt with the morality check AI's feedback and sends a new JSON request to the generative AI.

[0687] Server processing of final response

[0688] The server receives the final approved answer and formats it in an appropriate display format using an output processing module, specifically, formatting the answer in a data format such as HTML or JSON, converting it into a format that is easy for users to understand.

[0689] User-submitted answers

[0690] The server sends the final answer to the user's device as an HTTP response. Specifically, it returns the formatted data to the device again via the HTTPS protocol.

[0691] Show Answers

[0692] The device displays the received response to the user. Specifically, the web browser or application analyzes the received data and displays it in the appropriate location on the screen. The user can see the following text: "To adapt to a new environment, it is important to actively communicate with those around you and strengthen self-management. Don't worry, just take it easy and get used to it."

[0693] This system allows users to quickly and safely receive appropriate information corresponding to their emotions, without being aware of the process that the generative AI is generating.

[0694] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0695] Step 1:

[0696] The user inputs the prompt using their own device. Specifically, the user types "Please tell me how to adapt to the new environment" into the input field of the device's web browser or dedicated application, and clicks the send button. The input data is the prompt text in text format.

[0697] Step 2:

[0698] The terminal sends the entered prompt to the server as an HTTP request. Specifically, JSON format data containing the user input is sent via the HTTPS protocol. The input data is the user prompt, and the output data is the HTTP request sent to the server.

[0699] Step 3:

[0700] The server passes the received prompt to the emotion engine and analyzes the user's emotion. Specifically, the emotion engine (which includes a natural language processing toolkit) analyzes the text and determines from keywords and context that the user is feeling "anxiety." The input data is the received prompt, and the output data is JSON-formatted data containing the analyzed emotion information.

[0701] Step 4:

[0702] The server passes the prompt containing the analyzed emotional information to the generative AI. Specifically, new JSON data with the results of the emotion engine added is sent as input to the generative AI. The generative AI generates an answer based on this. The input data is the prompt containing emotional information, and the output data is JSON format data containing the generated answer.

[0703] Step 5:

[0704] The server receives the generated answer and then sends it to the morality check AI. The response from the generation AI comes back in JSON format, which the server parses. The input data is the generated answer, and the output data is the data sent to the morality check AI.

[0705] Step 6:

[0706] The server sends the generated answer to the morality check AI, which evaluates the appropriateness of the answer. Specifically, the morality check AI (a model with specific ethical filters) analyzes the answer and checks whether it contains inappropriate content or misinformation. If the morality check AI determines that the answer is appropriate, this information is also returned in JSON format. The input data is the generated answer, and the output data is data containing the evaluation result of the appropriateness.

[0707] Step 7:

[0708] The server receives the evaluation results from the morality check AI. If the answer is judged to be inappropriate, it sends a request to the generative AI again based on the correction suggestions provided by the morality check AI to generate a new answer. Specifically, it appropriately combines the original prompt and the morality check AI's feedback and sends a new JSON request to the generative AI. The input data is the evaluation result from the morality check AI, and the output data is the corrected prompt.

[0709] Step 8:

[0710] The server finally receives the approved answer and formats it in the appropriate display format in the output processing module. Specifically, the answer content is formatted in data formats such as HTML or JSON and converted into a form that is easy for users to understand. The input data is the approved answer, and the output data is the formatted answer.

[0711] Step 9:

[0712] The server sends the final answer to the user's terminal as an HTTP response. Specifically, it returns the formatted data to the terminal again via the HTTPS protocol. The input data is the formatted answer, and the output data is the HTTP response.

[0713] Step 10:

[0714] The device displays the received response to the user. Specifically, the web browser or application analyzes the received data and displays it in the appropriate location on the screen. The user can see the following text: "To adapt to a new environment, it is important to actively communicate with those around you and strengthen self-management. Don't worry, just take it easy and get used to it." The input data is the HTTP response, and the output data is the text displayed to the user.

[0715] (Application example 2)

[0716] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0717] Conventional food delivery systems simply present a menu without considering the user's emotions or mood when ordering. This results in a lack of flexible suggestions and advice based on the user's current mental state and emotions, leading to a lack of satisfaction in the user experience. Furthermore, the inability to provide appropriate information based on the user's emotions makes it difficult to provide the comfortable service that users desire.

[0718] 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.

[0719] In this invention, the server includes means for receiving a prompt input by a user, generative AI means for generating an answer based on the prompt, emotion analysis means for providing emotion analysis information to the generative AI means, morality check AI means for evaluating the appropriateness of the generated answer, and means for transmitting to a terminal recommendation information based on the answer approved by the morality check AI means and the emotion analysis information, thereby making it possible to provide the user with appropriate meal menu suggestions and advice in the form of customized messages according to their emotions and moods.

[0720] "User" refers to an individual or organization that uses this system.

[0721] A "prompt" is a request or question that a user enters into a system.

[0722] "Generative artificial intelligence means" refers to AI technologies that generate appropriate answers based on user prompts.

[0723] "Emotion analysis means" refers to a technique for analyzing emotion information from a prompt entered by a user.

[0724] "Moral check artificial intelligence means" refers to AI technology that evaluates the appropriateness of generated answers and eliminates inappropriate content.

[0725] "Terminal" refers to a device for providing generated answers and recommendation information to a user.

[0726] "Emotion analysis information" refers to data that indicates the emotional state of the user extracted by the emotion analysis means.

[0727] "Recommended information" refers to specific options or advice suggested to users based on sentiment analysis information.

[0728] The present invention relates to a system that generates answers to prompts entered by a user, evaluates the appropriateness of the answers, and provides appropriate information based on the user's emotions. In particular, the present invention is applied to food delivery, and provides appropriate menu suggestions and emotional support messages by taking into account the user's emotions when ordering a meal.

[0729] System configuration

[0730] The system consists of the following main components:

[0731] 1. Input Processing Module: A user uses a terminal to input a prompt, for example, the user inputs "I'm feeling down today."

[0732] 2. Generative AI methods: Generate appropriate answers based on user prompts. For example, "If you're looking for something to cheer you up, I recommend chicken soup or salad."

[0733] 3. Emotion analysis: Analyzes emotional information from user input. Example: Analyzes "I feel depressed."

[0734] 4. Moral Check AI Method: Evaluate the appropriateness of the generated answer. For example, check whether the generated answer is harmful to the user’s feelings.

[0735] 5. Output processing module: Sends the final answer and recommendation information to the terminal and displays it. Example: Display "If you're looking for something to cheer you up, we recommend chicken soup or salad."

[0736] Hardware and software used

[0737] Hardware: User devices (smartphones, tablets, etc.), servers

[0738] Software: Python, requests library, emotion analysis module, generative AI module, moral check module, menu recommendation module

[0739] Data processing and calculation

[0740] 1. User prompt input: The terminal receives the user prompt and sends it to the server as an HTTP request.

[0741] 2. Sentiment analysis: The server passes the prompt received to the sentiment analyzer, which analyzes the user's sentiment based on the context and keywords in the text. For example, "I feel depressed" is analyzed and determined to be "negative."

[0742] 3. Answer generation by generative AI: The prompt containing the analyzed emotional information is passed to a generative AI means to generate an answer. The generated answer is in line with the user's emotions. Example: "If you're looking for something to cheer you up, I recommend chicken soup or salad."

[0743] 4. Moral check: The generated answer is sent to the moral check AI means to evaluate its appropriateness. If it is judged to be inappropriate, a correction suggestion is regenerated. Example: The generated answer "Chicken soup or salad is recommended" is judged to be appropriate.

[0744] 5. Sending the final answer: The output processing module formats the accepted answer and sends it to the user's device as an HTTP response. The user's device displays the text "If you're looking for something to cheer you up, I recommend chicken soup or salad."

[0745] Specific examples

[0746] A user uses a food delivery assistant application and inputs, "I'm feeling down today." The device sends this to the server. The server receives the prompt, and the emotion analysis means analyzes the emotional information (negative). Based on the analyzed emotional information, the generative AI generates an answer: "If you're looking for something to cheer you up, I recommend chicken soup or salad." After the moral check AI determines this answer is appropriate, the final answer is sent to the user's device and displayed.

[0747] Example of a user prompt:

[0748] "I'm feeling down today, so I'd like you to recommend a dish that will cheer me up."

[0749] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0750] Step 1:

[0751] The user inputs a prompt through the terminal. Specifically, the user inputs "I'm feeling depressed today." This input text is acquired as prompt information.

[0752] Input: User prompt (e.g., "I'm feeling down today")

[0753] Output: Prompt information (text)

[0754] Step 2:

[0755] The terminal sends the input prompt information to the server as an HTTP request. Specifically, the prompt information is sent to the server's input processing module using a REST API or similar.

[0756] Input: Prompt information (text)

[0757] Output: Prompt information sent to the server (HTTP request)

[0758] Step 3:

[0759] The server passes the received prompt information to the emotion analysis means, which analyzes the user's emotion. Specifically, it uses a text analysis algorithm to recognize the user's emotion (e.g., "negative") from the prompt content.

[0760] Input: Prompt information (text)

[0761] Output: Sentiment analysis information (e.g., "negative")

[0762] Step 4:

[0763] The server passes prompt information, including the analyzed emotional information, to the generative AI means. The generative AI model generates an appropriate response based on the prompt information and emotional information. Specifically, in response to the statement "I'm feeling down today," the AI ​​model generates the response "If you're looking for something to cheer you up, I recommend chicken soup or a salad."

[0764] Input: prompt information (text), sentiment analysis information (e.g., "negative")

[0765] Output: Generated answer (e.g., "For comfort food, chicken soup or salad is recommended.")

[0766] Step 5:

[0767] The server sends the generated answer to the morality check AI means, which evaluates the appropriateness of the answer. Specifically, the morality check AI checks whether the generated answer contains inappropriate content or misinformation, and returns an evaluation result of whether the answer is appropriate.

[0768] Input: Generated answer (text)

[0769] Output: Evaluation result (appropriate / inappropriate)

[0770] Step 6:

[0771] The server receives the evaluation results from the moral check AI, and if the evaluation results are appropriate, it sends the final answer and recommendation information to the device. Specifically, it sends the answer that it judges to be appropriate (e.g., "If you're looking for something to cheer you up, we recommend chicken soup or salad.") to the user's device as an HTTP response.

[0772] Input: Evaluation results, generated answers (text)

[0773] Output: Final answer (text) sent to the terminal

[0774] Step 7:

[0775] The device displays the final answer it received to the user. Specifically, it displays "If you're looking for something to cheer you up, we recommend chicken soup or salad" in an easy-to-read format on the device's UI (user interface).

[0776] Input: Final answer (text)

[0777] Output: what is displayed to the user (text)

[0778] 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.

[0779] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0780] 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.

[0781] [Third embodiment]

[0782] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0783] 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.

[0784] 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).

[0785] 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.

[0786] 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.

[0787] 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).

[0788] 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.

[0789] 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.

[0790] 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.

[0791] 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.

[0792] 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.

[0793] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[0794] The present invention is a system that uses generative artificial intelligence (AI) to generate answers to prompts entered by users and evaluate the appropriateness of those answers, allowing users to receive safe and appropriate information.

[0795] System Configuration

[0796] The system consists of the following main components:

[0797] 1. Input Processing Module

[0798] 2. Generative AI

[0799] 3. Moral Check AI

[0800] 4. Output Processing Module

[0801] User prompt input

[0802] The user uses the terminal to enter a prompt, for example, "Tell me how to adapt to a new environment," and this prompt is sent from the terminal to the server.

[0803] Server-sent prompts

[0804] The device sends the input prompt to the server, which converts it into an appropriate format and passes it to the generative AI.

[0805] Answer generation using generative AI

[0806] The server passes the received prompt to the generative AI, which then generates an answer based on the prompt. For example, the generative AI might generate an answer such as, "In order to adapt to a new environment, it is important to actively communicate with those around you and strengthen self-management."

[0807] Server reception of response

[0808] The server receives the generated answer and then sends the answer to the morale check AI.

[0809] Evaluation by moral check AI

[0810] The server sends the generated answer to the morality check AI, which evaluates the appropriateness of the answer. The morality check AI uses natural language processing technology to check whether the answer contains inappropriate content or misinformation. For example, the morality check AI evaluates the answer and determines that it is appropriate.

[0811] Corrective action (if necessary)

[0812] The server receives the evaluation results from the moral check AI and sends the approved answer directly to the user. On the other hand, if the answer is judged to be inappropriate, it sends a request to the generative AI again based on the moral check AI's suggested corrections, and a new answer is generated.

[0813] Server processing of final response

[0814] The server receives the final approved response and formats it in an appropriate display format in the output processing module.

[0815] User-submitted answers

[0816] The server sends the final answer to the user's device, for example, by sending the answer data in an HTTP response.

[0817] Show Answers

[0818] The device displays the received answer to the user in an easy-to-read format on the device's interface (web browser, application, etc.).

[0819] Specific examples

[0820] For example:

[0821] The user types "Please tell me how to adapt to a new environment" into their device. The device sends this prompt to the server. The server passes the prompt to the generative AI, which generates an answer such as "In order to adapt to a new environment, it is important to actively communicate with those around you and strengthen self-management." The server passes the generated answer to the morality check AI, which evaluates the answer and determines it is appropriate. The server sends the approved answer to the user's device, and the device displays the answer to the user.

[0822] This system allows users to quickly receive appropriate information without being aware of the process by which the generative AI generates it.

[0823] The processing flow will be explained below.

[0824] Step 1:

[0825] The user types a prompt into the terminal, for example, "Tell me how to adapt to a new environment."

[0826] Step 2:

[0827] The terminal sends the entered prompt to the server as an HTTP request.

[0828] Step 3:

[0829] The server converts the received prompt into an appropriate format and sends a request to the generative AI API. For example, it converts it into JSON format and passes it to the generative AI.

[0830] Step 4:

[0831] The generative AI generates answers based on prompts received from the server, such as "To adapt to a new environment, it is important to actively communicate with those around you and strengthen self-management."

[0832] Step 5:

[0833] The server receives the answer generated by the generative AI.

[0834] Step 6:

[0835] The server sends the generated answer to the morality check AI, for example, by sending a request to the morality check AI's API to check whether the generated answer is appropriate.

[0836] Step 7:

[0837] The moral check AI analyzes the answers it receives using natural language processing technology and evaluates the appropriateness of the answers, for example, checking to see if they contain inappropriate language or misinformation.

[0838] Step 8:

[0839] The server receives the evaluation results from the moral check AI. Approved answers proceed to the next step, while answers deemed inappropriate are sent to the generative AI for revision, which then generates a new answer.

[0840] Step 9:

[0841] The server then formats the accepted response in an output processing module into an appropriate display format, for example, HTML or JSON.

[0842] Step 10:

[0843] The server sends the final answer to the user's device as an HTTP response.

[0844] Step 11:

[0845] The device displays the received answer to the user, for example, in a web browser or application in an easy-to-read format.

[0846] Example 1

[0847] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0848] Conventional systems using generative AI lack sufficient means to verify whether the answers to user-input instructions are appropriate, resulting in a high risk of providing users with inappropriate content or erroneous information. Furthermore, if an inappropriate answer is generated, there is no mechanism for automatically correcting it, meaning users end up receiving answers with low reliability. This makes it difficult for users to use the system with confidence.

[0849] 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.

[0850] In this invention, the server includes means for receiving a command input by a user, means for transmitting the command to the server, generative AI means for generating an answer based on the command, ethics check AI means for evaluating the appropriateness of the generated answer, means for transmitting the approved answer to the terminal, and means for the terminal to display the answer to the user, thereby enabling the user to quickly receive reliable and appropriate information.

[0851] "User" refers to a person who utilizes the system to input instructions and receive generated responses.

[0852] "Instruction text" refers to text data such as questions or requests that a user inputs into the system.

[0853] "Server" refers to a central computing unit that receives instructions, processes them, and sends the generated answers to the user's terminal.

[0854] "Terminal" refers to a device used by a user to input instructions and receive and display responses. Examples include a personal computer and a smartphone.

[0855] "Generative artificial intelligence means" refers to an artificial intelligence system or program that generates answers based on input instructions.

[0856] "Ethics check artificial intelligence means" refers to an artificial intelligence system or program that evaluates the appropriateness of generated answers and detects inappropriate content or misinformation.

[0857] "Formatting" refers to the process of converting generated answers into a user-readable form.

[0858] A "prompt" is synonymous with an instruction, and refers to a question or request that a user enters into a system.

[0859] This invention is a system that generates appropriate answers to user inputs, verifies the contents of the answers, and provides them to the user. This system consists of the following main components:

[0860] 1. Input Processing Module

[0861] 2. Generative Artificial Intelligence

[0862] 3. Ethics Checking AI

[0863] 4. Output Processing Module

[0864] First, the user inputs an instruction using the terminal. For example, if the user inputs "Tell me how to adapt to a new environment," this instruction is sent from the terminal to the server.

[0865] Specifically, the device sends the input instruction text to the server as an HTTP POST request. The server then passes the received instruction text to a generative AI (e.g., the OpenAI GPT series). The generative AI generates an answer based on the instruction text. For example, it might generate an answer such as, "In order to adapt to a new environment, it is important to actively communicate with those around you and strengthen self-management."

[0866] The server then sends the generated answers to an ethics check AI (e.g., a filtering model) that checks whether the generated answers contain inappropriate content or misinformation. This verification reduces the risk of inappropriate answers being provided to users.

[0867] If the generated answer is judged to be inappropriate, the server sends another instruction to the generative AI, requesting it to generate a new answer. This process is repeated until an appropriate answer is generated.

[0868] Finally, the server passes the accepted answer to the output processing module, which formats the answer in a format suitable for the user (for example, converting the generated text into HTML and formatting it so that it is easy for the user to read). Finally, the server sends the formatted answer as an HTTP response to the user's terminal.

[0869] The device then displays the received answers to the user through a web browser or dedicated application interface. This allows the user to quickly receive appropriate and reliable answers without being aware of the generative AI process.

[0870] As a concrete example, a user types "Tell me how to adapt to a new environment" into a terminal, and the terminal sends this to a server. The server passes the prompt to a generative AI, which generates an answer like the one above. The server then passes the generated answer to an ethics check AI for evaluation. If the ethics check AI approves the answer, the server sends it to the terminal, which then displays it to the user.

[0871] This enables systems that make full use of "generative AI models" and "prompt sentences" to provide users with safe and appropriate information.

[0872] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0873] Program processing steps

[0874] Step 1:

[0875] The user uses the terminal to input instructions.

[0876] Input: Text data such as "Tell me how to adapt to a new environment."

[0877] What happens: A user enters text into an input field using a keyboard or voice recognition.

[0878] Step 2:

[0879] The terminal sends the input instruction to the server.

[0880] Input: Instruction text data.

[0881] Output: The HTTP POST request sent to the server.

[0882] Specific behavior: The input text is included in the body of the HTTP request and sent to the specified URL.

[0883] Step 3:

[0884] The server receives the instructions and passes them to the generative artificial intelligence.

[0885] Input: The HTTP POST request sent from the device.

[0886] Output: API request to the generative AI.

[0887] Specific operation: The server extracts instructions from the body of the HTTP request and sends the request to the API of the generative artificial intelligence (e.g., OpenAI GPT series).

[0888] Step 4:

[0889] Generative artificial intelligence generates answers based on instructions.

[0890] Input: Instruction text data.

[0891] Output: The generated answer text.

[0892] How it works: Generative AI uses natural language processing to generate appropriate answers based on the user's instructions. For example, it generates an answer such as, "In order to adapt to a new environment, it is important to actively communicate with those around you and strengthen self-management."

[0893] Step 5:

[0894] The server receives the generated answer and sends it to the ethical check artificial intelligence.

[0895] Input: Answer text received from the generative AI.

[0896] Output: API request to the Ethics Check AI.

[0897] Specific operation: The server extracts the generated answers and sends a request to the API of the ethical check artificial intelligence (e.g., filtering model).

[0898] Step 6:

[0899] Ethics check: Artificial intelligence evaluates the appropriateness of generated answers.

[0900] Input: Answer text data.

[0901] Output: The result of evaluating whether the answer is correct.

[0902] What it does: Ethics check AI analyzes your answers and checks for inappropriate content or misinformation.

[0903] Step 7:

[0904] The server receives the evaluation results of the ethical check artificial intelligence and, if necessary, sends a regeneration request to the generative artificial intelligence.

[0905] Input: Evaluation results from Ethics Check Artificial Intelligence.

[0906] Output: Repeat request to generative AI if necessary.

[0907] Specific operation: If the answer is determined to be inappropriate, the server sends a regeneration request to the generative artificial intelligence to generate a new answer.

[0908] Step 8:

[0909] The server formats the accepted response in the output processing module.

[0910] Input: The accepted answer text.

[0911] Output: The formatted answer text.

[0912] Specific behavior: Converts the accepted answer text into a user-readable format (e.g., HTML format).

[0913] Step 9:

[0914] The server sends the formatted response to the user's terminal.

[0915] Input: The formatted answer text.

[0916] Output: The HTTP response sent to the user's device.

[0917] Specific behavior: Sends formatted text to the terminal as an HTTP response.

[0918] Step 10:

[0919] The terminal displays the received answer to the user.

[0920] Input: The HTTP response received from the server.

[0921] Output: The answer text that is displayed on the screen.

[0922] Specific operation: Display the received data in an easy-to-read format in a web browser or application interface.

[0923] Through these steps, users can receive safe and relevant information quickly.

[0924] (Application example 1)

[0925] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0926] In factories, workers often have questions or problems regarding production lines, machine operation, and quality control, and it is difficult to obtain appropriate answers each time. This reduces work efficiency and increases the risk of operational errors. Furthermore, utilizing human expertise requires a great deal of time and effort.

[0927] 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.

[0928] In this invention, the server includes means for receiving prompts input by a user, generative AI means for generating answers based on the prompts, morality check AI means for evaluating the appropriateness of the generated answers, means for transmitting answers approved by the morality check AI means to an output device, and means for providing answers to questions from factory workers through the output device, thereby making it possible to provide factory workers with prompt and appropriate information in response to their doubts and problems.

[0929] A "user" is a person who uses a system or device and is a worker who is responsible for inputting prompts.

[0930] A "prompt" refers to a question or instruction entered by a user, and is input information used by generative artificial intelligence to generate an answer.

[0931] "Generative AI" is an AI technology that generates appropriate answers based on prompts entered by the user.

[0932] "Moral check AI" is an AI technology that evaluates whether the generated answers are ethically appropriate.

[0933] "Output device" refers to a device for providing the user with the final answer generated by the generative artificial intelligence and the moral check artificial intelligence.

[0934] "Factory worker" means a person who operates production lines and machines, performs quality control, and other tasks within a factory.

[0935] A "question" is a question or inquiry for information that a factory worker has, and is entered as a prompt.

[0936] An "answer" is information generated by the generative AI based on a prompt and provided after being evaluated by the moral check AI.

[0937] This invention is a system that provides prompt and appropriate information to factory workers in response to questions and problems. This system utilizes generative AI and moral check AI to efficiently generate answers and evaluate their appropriateness.

[0938] The system is structured as follows: First, a user inputs a question for a factory worker as a prompt. In response, the input prompt is sent to the server. The server uses generative AI to generate an answer to the prompt. The generated answer is then evaluated for appropriateness by morality check AI.

[0939] The answers that have been approved are sent from the server to an output device and provided to the factory workers. If they are deemed inappropriate, they are regenerated or corrected. For example, OpenAI's API is used for generative AI. Dedicated natural language processing technology is used for moral check AI.

[0940] The specific system operation procedure is as follows: A factory worker speaks to the robot, saying, "Please tell me how to operate this machine." The robot receives this prompt and sends it to the server. The generative AI generates an answer such as, "To operate this machine, first turn it on, then follow the setup guide and press the buttons." The appropriateness of the answer is evaluated by the moral check AI, and if it is deemed appropriate, the robot provides the answer to the factory worker.

[0941] This system allows factory workers to efficiently obtain the appropriate information. Below are some example prompts:

[0942] Example prompt:

[0943] 1. "How do I set up a new production line?"

[0944] 2. "Where are the quality checkpoints for this product?"

[0945] 3. "Please explain how to deal with mechanical problems."

[0946] This will improve work efficiency within the factory and reduce the risk of operational errors and problems.

[0947] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0948] Step 1: The user enters the prompt.

[0949] Input: A factory worker speaks a question to the robot.

[0950] Output: The robot receives the user's prompt and converts it into a data format.

[0951] Specific operation: When a factory worker speaks to the robot, saying, "Please tell me how to operate this machine," the robot's voice recognition function converts the question into text data.

[0952] Step 2: Send the entered prompt from the terminal to the server.

[0953] Input: The prompt converted into text data by the robot.

[0954] Output: Text data is sent to the server.

[0955] Specific operation: The robot converts the prompt text data into an HTTP request format and sends it to the server.

[0956] Step 3: The server generates an answer using generative artificial intelligence.

[0957] Input: The prompt data sent to the server.

[0958] Output: Answer text generated by generative artificial intelligence.

[0959] Specific operation: The server passes the received prompt data to the OpenAI API, and the generative AI model generates an answer such as, "To operate this machine, first turn it on, then press the buttons according to the setup guide."

[0960] Step 4: The generated answers are sent to the moral check AI for evaluation.

[0961] Input: Answer text generated by generative artificial intelligence.

[0962] Output: Evaluation result (approval or disapproval) by the moral check AI.

[0963] Specific operation: The server sends the generated answer to the moral check AI, which uses natural language processing technology to evaluate the appropriateness of the answer and check whether it contains inappropriate content or misinformation.

[0964] Step 5: Revise or regenerate answers based on the results of the moral check AI.

[0965] Input: The generated answer when the evaluation result of the moral check artificial intelligence is disapproval.

[0966] Output: The corrected or regenerated answer text.

[0967] Specific operation: If the moral check AI judges the generated answer to be inappropriate, the server will again prompt the generator AI to generate a new answer. It may also request a regeneration by adding a phrase such as "Please try again."

[0968] Step 6: Moral Check The answer approved by the AI ​​is sent to the terminal.

[0969] Input: Answer text approved by the Morality Check AI.

[0970] Output: The accepted answer is sent to the robot.

[0971] Specific operation: The server sends the final accepted answer to the robot as an HTTP response.

[0972] Step 7: The robot provides the answer to the factory worker.

[0973] Input: The accepted answer text received from the server.

[0974] Output: Provides information to factory workers.

[0975] Specific operation: The robot uses a voice synthesis function to audibly convey the received response to the factory worker. For example, it might say something like, "To operate this machine, first turn it on, then follow the setup guide and press the buttons."

[0976] Through the above steps, factory workers can efficiently obtain appropriate information, thereby improving the accuracy and efficiency of their work.

[0977] 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.

[0978] This invention uses generative artificial intelligence (AI) to generate answers to prompts entered by users, evaluates the appropriateness of those answers, and combines this with an emotion engine that recognizes the user's emotions to provide more accurate and appropriate information. This system allows users to receive safe and appropriate information in a style appropriate to their emotions.

[0979] System Configuration

[0980] The system consists of the following main components:

[0981] 1. Input Processing Module

[0982] 2. Generative AI

[0983] 3. Moral Check AI

[0984] 4. Emotion Engine

[0985] 5. Output Processing Module

[0986] User prompt input

[0987] The user uses the terminal to enter a prompt, for example, "Tell me how to adapt to a new environment," and this prompt is sent from the terminal to the server.

[0988] Server-sent prompts

[0989] The terminal sends the entered prompt to the server as an HTTP request.

[0990] Emotion Analysis

[0991] The server passes the received prompt to the emotion engine, which analyzes the user's emotion. For example, it analyzes that the user is feeling "anxiety" based on the context and keywords of the text.

[0992] Answer generation using generative AI

[0993] The server passes the prompt, including the analyzed emotional information, to the generative AI. The generative AI generates an answer based on the prompt and the emotional information. For example, it generates an answer such as, "To adapt to a new environment, it is important to actively communicate with those around you and strengthen self-management. Don't worry, just take your time to get used to it."

[0994] Server reception of response

[0995] The server receives the generated answer and then sends the answer to the morale check AI.

[0996] Evaluation by moral check AI

[0997] The server sends the generated answer to the morality check AI, which evaluates the appropriateness of the answer. The morality check AI uses natural language processing technology to check whether the answer contains inappropriate content or misinformation. For example, the morality check AI evaluates the answer and determines that it is appropriate.

[0998] Corrective action (if necessary)

[0999] The server receives the evaluation results from the moral check AI and sends the approved answer directly to the user. On the other hand, if the answer is judged to be inappropriate, it sends a request to the generative AI again based on the moral check AI's suggested corrections, and a new answer is generated.

[1000] Server processing of final response

[1001] The server receives the final approved response and formats it in an appropriate display format in the output processing module.

[1002] User-submitted answers

[1003] The server sends the final answer to the user's terminal as an HTTP response.

[1004] Show Answers

[1005] The terminal displays the received answer to the user in an easy-to-read format on the terminal interface (web browser, application, etc.).

[1006] Specific examples

[1007] For example:

[1008] The user types "Tell me how to adapt to a new environment" into their device. The device sends this prompt to the server. The server passes the prompt to the emotion engine, which analyzes the user's emotion as "anxiety." The server passes the prompt along with the analyzed emotion information to the generative AI. The generative AI generates an answer: "To adapt to a new environment, it's important to actively communicate with those around you and strengthen self-management. Don't worry, just get used to it slowly." The server passes the generated answer to the morality check AI, which evaluates the answer and determines it is appropriate. The server sends the approved answer to the user's device, and the device displays the answer to the user.

[1009] This system allows users to quickly and safely receive appropriate information corresponding to their emotions, without being aware of the process that the generative AI is generating.

[1010] The processing flow will be explained below.

[1011] Step 1:

[1012] The user types a prompt into the terminal, for example, "Tell me how to adapt to a new environment."

[1013] Step 2:

[1014] The terminal sends the entered prompt to the server as an HTTP request.

[1015] Step 3:

[1016] The server passes the received prompt to the emotion engine, which analyzes the user's emotion. For example, it analyzes that the user is feeling "anxiety" based on the context and keywords of the text.

[1017] Step 4:

[1018] The server passes the prompt containing the analyzed emotional information to the generative AI. For example, the emotional information "anxiety" is sent along with the prompt.

[1019] Step 5:

[1020] Generative AI generates answers that take emotional information into account, such as, "To adapt to a new environment, it's important to actively communicate with those around you and strengthen self-management. Don't worry, just take your time to get used to it."

[1021] Step 6:

[1022] The server receives the answer generated by the generative AI.

[1023] Step 7:

[1024] The server sends the generated answer to the morality check AI, for example, by sending a request to the morality check AI's API to check whether the generated answer is appropriate.

[1025] Step 8:

[1026] The moral check AI analyzes the answers it receives using natural language processing technology and evaluates the appropriateness of the answers, for example, checking to see if they contain inappropriate language or misinformation.

[1027] Step 9:

[1028] The server receives the evaluation results from the moral check AI. Approved answers proceed to the next step, while answers deemed inappropriate are sent to the generative AI for revision, which then generates a new answer.

[1029] Step 10:

[1030] The server then formats the accepted response in an output processing module into an appropriate display format, for example, HTML or JSON.

[1031] Step 11:

[1032] The server sends the final answer to the user's device as an HTTP response.

[1033] Step 12:

[1034] The device will display the received answer to the user. For example, the answer will be displayed in an easy-to-read format in a web browser or application. Specifically, the device will display a message such as, "To adapt to a new environment, it is important to actively communicate with those around you and strengthen self-management. Don't worry, just take your time to get used to it."

[1035] Example 2

[1036] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1037] Conventional generative AI systems generate answers that do not sufficiently consider the user's emotions, which can result in the system failing to provide the appropriate information the user is looking for. Furthermore, there is a lack of a process for determining the appropriateness of the generated answers, which creates a risk of providing inappropriate information to the user. As a result, there is a risk of the system losing user trust.

[1038] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes means for receiving a prompt input by a user, emotion engine means for analyzing the prompt and recognizing the user's emotion, generative AI means for generating an answer based on the prompt including the analyzed emotion information, morality check AI means for evaluating the appropriateness of the generated answer, means for transmitting the answer approved by the morality check AI means to a terminal, and means for displaying the answer on the terminal. This makes it possible to provide appropriate and safe information according to the user's emotion.

[1039] A "prompt" is text data such as a question or instruction entered by a user.

[1040] An "emotion engine" is software or hardware that analyzes a user's prompts and recognizes the user's emotions from their context and keywords.

[1041] "Generative AI" is an AI model for generating answers in natural language based on prompts and emotional information.

[1042] "Moral check AI" is an AI model that evaluates the appropriateness of answers generated by generative AI and checks whether they contain inappropriate content or misinformation.

[1043] A "terminal" is a device (such as a PC or smartphone) on which a user enters prompts and receives generated answers.

[1044] This invention combines a system that uses generative artificial intelligence (AI) to generate answers to prompts entered by users and evaluates whether the answers are appropriate, with an emotion engine that recognizes the user's emotions. This system allows users to receive safe and appropriate information in a style appropriate to their emotions.

[1045] System Configuration

[1046] The system consists of the following main components:

[1047] 1. Input Processing Module

[1048] 2. Generative AI

[1049] 3. Moral Check AI

[1050] 4. Emotion Engine

[1051] 5. Output Processing Module

[1052] User prompt input

[1053] The user uses the terminal to enter a prompt. Specifically, the user types, "Tell me how to adapt to a new environment." This prompt is sent from the terminal to the server.

[1054] Server-sent prompts

[1055] The terminal sends the entered prompt to the server as an HTTP request.

[1056] Emotion Analysis

[1057] The server passes the received prompt to the emotion engine, which analyzes the user's emotion. The emotion engine, for example, uses a natural language processing toolkit to analyze the context and keywords in the text to determine that the user is feeling "anxiety." This information is converted into JSON format and passed on to the next process.

[1058] Answer generation using generative AI

[1059] The server passes the prompt, including the analyzed emotional information, to the generative AI. Specifically, new JSON data with the emotion engine results added is sent as input to the generative AI. The generative AI then generates an answer based on the prompt and emotional information, using, for example, OpenAI's GPT model. An example of a generated answer would be, "To adapt to a new environment, it's important to actively communicate with those around you and strengthen self-management. Don't worry, just take it easy and get used to it."

[1060] Server reception of response

[1061] The server receives the generated answer and then sends it to the moral check AI. The response from the generating AI comes back in JSON format, which the server parses.

[1062] Evaluation by moral check AI

[1063] The server sends the generated answers to a morality check AI, which evaluates their appropriateness. The morality check AI analyzes the answers created by the generative AI, for example, using a model with specific ethical filters, to check for inappropriate content or misinformation. If the morality check AI determines that the answers are appropriate, this information is also returned in JSON format.

[1064] Corrective action (if necessary)

[1065] The server receives the evaluation results from the morality check AI. If the answer is judged to be inappropriate, it sends a request to the generative AI again based on the correction suggestions provided by the morality check AI to generate a new answer. Specifically, it appropriately combines the original prompt with the morality check AI's feedback and sends a new JSON request to the generative AI.

[1066] Server processing of final response

[1067] The server receives the final approved answer and formats it in an appropriate display format using an output processing module, specifically, formatting the answer in a data format such as HTML or JSON, converting it into a format that is easy for users to understand.

[1068] User-submitted answers

[1069] The server sends the final answer to the user's device as an HTTP response. Specifically, it returns the formatted data to the device again via the HTTPS protocol.

[1070] Show Answers

[1071] The device displays the received response to the user. Specifically, the web browser or application analyzes the received data and displays it in the appropriate location on the screen. The user can see the following text: "To adapt to a new environment, it is important to actively communicate with those around you and strengthen self-management. Don't worry, just take it easy and get used to it."

[1072] This system allows users to quickly and safely receive appropriate information corresponding to their emotions, without being aware of the process that the generative AI is generating.

[1073] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1074] Step 1:

[1075] The user inputs the prompt using their own device. Specifically, the user types "Please tell me how to adapt to the new environment" into the input field of the device's web browser or dedicated application, and clicks the send button. The input data is the prompt text in text format.

[1076] Step 2:

[1077] The terminal sends the entered prompt to the server as an HTTP request. Specifically, JSON format data containing the user input is sent via the HTTPS protocol. The input data is the user prompt, and the output data is the HTTP request sent to the server.

[1078] Step 3:

[1079] The server passes the received prompt to the emotion engine and analyzes the user's emotion. Specifically, the emotion engine (which includes a natural language processing toolkit) analyzes the text and determines from keywords and context that the user is feeling "anxiety." The input data is the received prompt, and the output data is JSON-formatted data containing the analyzed emotion information.

[1080] Step 4:

[1081] The server passes the prompt containing the analyzed emotional information to the generative AI. Specifically, new JSON data with the results of the emotion engine added is sent as input to the generative AI. The generative AI generates an answer based on this. The input data is the prompt containing emotional information, and the output data is JSON format data containing the generated answer.

[1082] Step 5:

[1083] The server receives the generated answer and then sends it to the morality check AI. The response from the generation AI comes back in JSON format, which the server parses. The input data is the generated answer, and the output data is the data sent to the morality check AI.

[1084] Step 6:

[1085] The server sends the generated answer to the morality check AI, which evaluates the appropriateness of the answer. Specifically, the morality check AI (a model with specific ethical filters) analyzes the answer and checks whether it contains inappropriate content or misinformation. If the morality check AI determines that the answer is appropriate, this information is also returned in JSON format. The input data is the generated answer, and the output data is data containing the evaluation result of the appropriateness.

[1086] Step 7:

[1087] The server receives the evaluation results from the morality check AI. If the answer is judged to be inappropriate, it sends a request to the generative AI again based on the correction suggestions provided by the morality check AI to generate a new answer. Specifically, it appropriately combines the original prompt and the morality check AI's feedback and sends a new JSON request to the generative AI. The input data is the evaluation result from the morality check AI, and the output data is the corrected prompt.

[1088] Step 8:

[1089] The server finally receives the approved answer and formats it in the appropriate display format in the output processing module. Specifically, the answer content is formatted in data formats such as HTML or JSON and converted into a form that is easy for users to understand. The input data is the approved answer, and the output data is the formatted answer.

[1090] Step 9:

[1091] The server sends the final answer to the user's terminal as an HTTP response. Specifically, it returns the formatted data to the terminal again via the HTTPS protocol. The input data is the formatted answer, and the output data is the HTTP response.

[1092] Step 10:

[1093] The device displays the received response to the user. Specifically, the web browser or application analyzes the received data and displays it in the appropriate location on the screen. The user can see the following text: "To adapt to a new environment, it is important to actively communicate with those around you and strengthen self-management. Don't worry, just take it easy and get used to it." The input data is the HTTP response, and the output data is the text displayed to the user.

[1094] (Application example 2)

[1095] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1096] Conventional food delivery systems simply present a menu without considering the user's emotions or mood when ordering. This results in a lack of flexible suggestions and advice based on the user's current mental state and emotions, leading to a lack of satisfaction in the user experience. Furthermore, the inability to provide appropriate information based on the user's emotions makes it difficult to provide the comfortable service that users desire.

[1097] 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.

[1098] In this invention, the server includes means for receiving a prompt input by a user, generative AI means for generating an answer based on the prompt, emotion analysis means for providing emotion analysis information to the generative AI means, morality check AI means for evaluating the appropriateness of the generated answer, and means for transmitting to a terminal recommendation information based on the answer approved by the morality check AI means and the emotion analysis information, thereby making it possible to provide the user with appropriate meal menu suggestions and advice in the form of customized messages according to their emotions and moods.

[1099] "User" refers to an individual or organization that uses this system.

[1100] A "prompt" is a request or question that a user enters into a system.

[1101] "Generative artificial intelligence means" refers to AI technologies that generate appropriate answers based on user prompts.

[1102] "Emotion analysis means" refers to a technique for analyzing emotion information from a prompt entered by a user.

[1103] "Moral check artificial intelligence means" refers to AI technology that evaluates the appropriateness of generated answers and eliminates inappropriate content.

[1104] "Terminal" refers to a device for providing generated answers and recommendation information to a user.

[1105] "Emotion analysis information" refers to data that indicates the emotional state of the user extracted by the emotion analysis means.

[1106] "Recommended information" refers to specific options or advice suggested to users based on sentiment analysis information.

[1107] The present invention relates to a system that generates answers to prompts entered by a user, evaluates the appropriateness of the answers, and provides appropriate information based on the user's emotions. In particular, the present invention is applied to food delivery, and provides appropriate menu suggestions and emotional support messages by taking into account the user's emotions when ordering a meal.

[1108] System configuration

[1109] The system consists of the following main components:

[1110] 1. Input Processing Module: A user uses a terminal to input a prompt, for example, the user inputs "I'm feeling down today."

[1111] 2. Generative AI methods: Generate appropriate answers based on user prompts. For example, "If you're looking for something to cheer you up, I recommend chicken soup or salad."

[1112] 3. Emotion analysis: Analyzes emotional information from user input. Example: Analyzes "I feel depressed."

[1113] 4. Moral Check AI Method: Evaluate the appropriateness of the generated answer. For example, check whether the generated answer is harmful to the user’s feelings.

[1114] 5. Output processing module: Sends the final answer and recommendation information to the terminal and displays it. Example: Display "If you're looking for something to cheer you up, we recommend chicken soup or salad."

[1115] Hardware and software used

[1116] Hardware: User devices (smartphones, tablets, etc.), servers

[1117] Software: Python, requests library, emotion analysis module, generative AI module, moral check module, menu recommendation module

[1118] Data processing and calculation

[1119] 1. User prompt input: The terminal receives the user prompt and sends it to the server as an HTTP request.

[1120] 2. Sentiment analysis: The server passes the prompt received to the sentiment analyzer, which analyzes the user's sentiment based on the context and keywords in the text. For example, "I feel depressed" is analyzed and determined to be "negative."

[1121] 3. Answer generation by generative AI: The prompt containing the analyzed emotional information is passed to a generative AI means to generate an answer. The generated answer is in line with the user's emotions. Example: "If you're looking for something to cheer you up, I recommend chicken soup or salad."

[1122] 4. Moral check: The generated answer is sent to the moral check AI means to evaluate its appropriateness. If it is judged to be inappropriate, a correction suggestion is regenerated. Example: The generated answer "Chicken soup or salad is recommended" is judged to be appropriate.

[1123] 5. Sending the final answer: The output processing module formats the accepted answer and sends it to the user's device as an HTTP response. The user's device displays the text "If you're looking for something to cheer you up, I recommend chicken soup or salad."

[1124] Specific examples

[1125] A user uses a food delivery assistant application and inputs, "I'm feeling down today." The device sends this to the server. The server receives the prompt, and the emotion analysis means analyzes the emotional information (negative). Based on the analyzed emotional information, the generative AI generates an answer: "If you're looking for something to cheer you up, I recommend chicken soup or salad." After the moral check AI determines this answer is appropriate, the final answer is sent to the user's device and displayed.

[1126] Example of a user prompt:

[1127] "I'm feeling down today, so I'd like you to recommend a dish that will cheer me up."

[1128] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1129] Step 1:

[1130] The user inputs a prompt through the terminal. Specifically, the user inputs "I'm feeling depressed today." This input text is acquired as prompt information.

[1131] Input: User prompt (e.g., "I'm feeling down today")

[1132] Output: Prompt information (text)

[1133] Step 2:

[1134] The terminal sends the input prompt information to the server as an HTTP request. Specifically, the prompt information is sent to the server's input processing module using a REST API or similar.

[1135] Input: Prompt information (text)

[1136] Output: Prompt information sent to the server (HTTP request)

[1137] Step 3:

[1138] The server passes the received prompt information to the emotion analysis means, which analyzes the user's emotion. Specifically, it uses a text analysis algorithm to recognize the user's emotion (e.g., "negative") from the prompt content.

[1139] Input: Prompt information (text)

[1140] Output: Sentiment analysis information (e.g., "negative")

[1141] Step 4:

[1142] The server passes prompt information, including the analyzed emotional information, to the generative AI means. The generative AI model generates an appropriate response based on the prompt information and emotional information. Specifically, in response to the statement "I'm feeling down today," the AI ​​model generates the response "If you're looking for something to cheer you up, I recommend chicken soup or a salad."

[1143] Input: prompt information (text), sentiment analysis information (e.g., "negative")

[1144] Output: Generated answer (e.g., "For comfort food, chicken soup or salad is recommended.")

[1145] Step 5:

[1146] The server sends the generated answer to the morality check AI means, which evaluates the appropriateness of the answer. Specifically, the morality check AI checks whether the generated answer contains inappropriate content or misinformation, and returns an evaluation result of whether the answer is appropriate.

[1147] Input: Generated answer (text)

[1148] Output: Evaluation result (appropriate / inappropriate)

[1149] Step 6:

[1150] The server receives the evaluation results from the moral check AI, and if the evaluation results are appropriate, it sends the final answer and recommendation information to the device. Specifically, it sends the answer that it judges to be appropriate (e.g., "If you're looking for something to cheer you up, we recommend chicken soup or salad.") to the user's device as an HTTP response.

[1151] Input: Evaluation results, generated answers (text)

[1152] Output: Final answer (text) sent to the terminal

[1153] Step 7:

[1154] The device displays the final answer it received to the user. Specifically, it displays "If you're looking for something to cheer you up, we recommend chicken soup or salad" in an easy-to-read format on the device's UI (user interface).

[1155] Input: Final answer (text)

[1156] Output: what is displayed to the user (text)

[1157] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1158] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1159] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1160] [Fourth embodiment]

[1161] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1162] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1163] 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).

[1164] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1165] 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.

[1166] 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).

[1167] 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.

[1168] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1169] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1170] 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.

[1171] 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.

[1172] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1173] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1174] The present invention is a system that uses generative artificial intelligence (AI) to generate answers to prompts entered by users and evaluate the appropriateness of those answers, allowing users to receive safe and appropriate information.

[1175] System Configuration

[1176] The system consists of the following main components:

[1177] 1. Input Processing Module

[1178] 2. Generative AI

[1179] 3. Moral Check AI

[1180] 4. Output Processing Module

[1181] User prompt input

[1182] The user uses the terminal to enter a prompt, for example, "Tell me how to adapt to a new environment," and this prompt is sent from the terminal to the server.

[1183] Server-sent prompts

[1184] The device sends the input prompt to the server, which converts it into an appropriate format and passes it to the generative AI.

[1185] Answer generation using generative AI

[1186] The server passes the received prompt to the generative AI, which then generates an answer based on the prompt. For example, the generative AI might generate an answer such as, "In order to adapt to a new environment, it is important to actively communicate with those around you and strengthen self-management."

[1187] Server reception of response

[1188] The server receives the generated answer and then sends the answer to the morale check AI.

[1189] Evaluation by moral check AI

[1190] The server sends the generated answer to the morality check AI, which evaluates the appropriateness of the answer. The morality check AI uses natural language processing technology to check whether the answer contains inappropriate content or misinformation. For example, the morality check AI evaluates the answer and determines that it is appropriate.

[1191] Corrective action (if necessary)

[1192] The server receives the evaluation results from the moral check AI and sends the approved answer directly to the user. On the other hand, if the answer is judged to be inappropriate, it sends a request to the generative AI again based on the moral check AI's suggested corrections, and a new answer is generated.

[1193] Server processing of final response

[1194] The server receives the final approved response and formats it in an appropriate display format in the output processing module.

[1195] User-submitted answers

[1196] The server sends the final answer to the user's device, for example, by sending the answer data in an HTTP response.

[1197] Show Answers

[1198] The device displays the received answer to the user in an easy-to-read format on the device's interface (web browser, application, etc.).

[1199] Specific examples

[1200] For example:

[1201] The user types "Please tell me how to adapt to a new environment" into their device. The device sends this prompt to the server. The server passes the prompt to the generative AI, which generates an answer such as "In order to adapt to a new environment, it is important to actively communicate with those around you and strengthen self-management." The server passes the generated answer to the morality check AI, which evaluates the answer and determines it is appropriate. The server sends the approved answer to the user's device, and the device displays the answer to the user.

[1202] This system allows users to quickly receive appropriate information without being aware of the process by which the generative AI generates it.

[1203] The processing flow will be explained below.

[1204] Step 1:

[1205] The user types a prompt into the terminal, for example, "Tell me how to adapt to a new environment."

[1206] Step 2:

[1207] The terminal sends the entered prompt to the server as an HTTP request.

[1208] Step 3:

[1209] The server converts the received prompt into an appropriate format and sends a request to the generative AI API. For example, it converts it into JSON format and passes it to the generative AI.

[1210] Step 4:

[1211] The generative AI generates answers based on prompts received from the server, such as "To adapt to a new environment, it is important to actively communicate with those around you and strengthen self-management."

[1212] Step 5:

[1213] The server receives the answer generated by the generative AI.

[1214] Step 6:

[1215] The server sends the generated answer to the morality check AI, for example, by sending a request to the morality check AI's API to check whether the generated answer is appropriate.

[1216] Step 7:

[1217] The moral check AI analyzes the answers it receives using natural language processing technology and evaluates the appropriateness of the answers, for example, checking to see if they contain inappropriate language or misinformation.

[1218] Step 8:

[1219] The server receives the evaluation results from the moral check AI. Approved answers proceed to the next step, while answers deemed inappropriate are sent to the generative AI for revision, which then generates a new answer.

[1220] Step 9:

[1221] The server then formats the accepted response in an output processing module into an appropriate display format, for example, HTML or JSON.

[1222] Step 10:

[1223] The server sends the final answer to the user's device as an HTTP response.

[1224] Step 11:

[1225] The device displays the received answer to the user, for example, in a web browser or application in an easy-to-read format.

[1226] Example 1

[1227] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1228] Conventional systems using generative AI lack sufficient means to verify whether the answers to user-input instructions are appropriate, resulting in a high risk of providing users with inappropriate content or erroneous information. Furthermore, if an inappropriate answer is generated, there is no mechanism for automatically correcting it, meaning users end up receiving answers with low reliability. This makes it difficult for users to use the system with confidence.

[1229] 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.

[1230] In this invention, the server includes means for receiving a command input by a user, means for transmitting the command to the server, generative AI means for generating an answer based on the command, ethics check AI means for evaluating the appropriateness of the generated answer, means for transmitting the approved answer to the terminal, and means for the terminal to display the answer to the user, thereby enabling the user to quickly receive reliable and appropriate information.

[1231] "User" refers to a person who utilizes the system to input instructions and receive generated responses.

[1232] "Instruction text" refers to text data such as questions or requests that a user inputs into the system.

[1233] "Server" refers to a central computing unit that receives instructions, processes them, and sends the generated answers to the user's terminal.

[1234] "Terminal" refers to a device used by a user to input instructions and receive and display responses. Examples include a personal computer and a smartphone.

[1235] "Generative artificial intelligence means" refers to an artificial intelligence system or program that generates answers based on input instructions.

[1236] "Ethics check artificial intelligence means" refers to an artificial intelligence system or program that evaluates the appropriateness of generated answers and detects inappropriate content or misinformation.

[1237] "Formatting" refers to the process of converting generated answers into a user-readable form.

[1238] A "prompt" is synonymous with an instruction, and refers to a question or request that a user enters into a system.

[1239] This invention is a system that generates appropriate answers to user inputs, verifies the contents of the answers, and provides them to the user. This system consists of the following main components:

[1240] 1. Input Processing Module

[1241] 2. Generative Artificial Intelligence

[1242] 3. Ethics Checking AI

[1243] 4. Output Processing Module

[1244] First, the user inputs an instruction using the terminal. For example, if the user inputs "Tell me how to adapt to a new environment," this instruction is sent from the terminal to the server.

[1245] Specifically, the device sends the input instruction text to the server as an HTTP POST request. The server then passes the received instruction text to a generative AI (e.g., the OpenAI GPT series). The generative AI generates an answer based on the instruction text. For example, it might generate an answer such as, "In order to adapt to a new environment, it is important to actively communicate with those around you and strengthen self-management."

[1246] The server then sends the generated answers to an ethics check AI (e.g., a filtering model) that checks whether the generated answers contain inappropriate content or misinformation. This verification reduces the risk of inappropriate answers being provided to users.

[1247] If the generated answer is judged to be inappropriate, the server sends another instruction to the generative AI, requesting it to generate a new answer. This process is repeated until an appropriate answer is generated.

[1248] Finally, the server passes the accepted answer to the output processing module, which formats the answer in a format suitable for the user (for example, converting the generated text into HTML and formatting it so that it is easy for the user to read). Finally, the server sends the formatted answer as an HTTP response to the user's terminal.

[1249] The device then displays the received answers to the user through a web browser or dedicated application interface. This allows the user to quickly receive appropriate and reliable answers without being aware of the generative AI process.

[1250] As a concrete example, a user types "Tell me how to adapt to a new environment" into a terminal, and the terminal sends this to a server. The server passes the prompt to a generative AI, which generates an answer like the one above. The server then passes the generated answer to an ethics check AI for evaluation. If the ethics check AI approves the answer, the server sends it to the terminal, which then displays it to the user.

[1251] This enables systems that make full use of "generative AI models" and "prompt sentences" to provide users with safe and appropriate information.

[1252] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1253] Program processing steps

[1254] Step 1:

[1255] The user uses the terminal to input instructions.

[1256] Input: Text data such as "Tell me how to adapt to a new environment."

[1257] What happens: A user enters text into an input field using a keyboard or voice recognition.

[1258] Step 2:

[1259] The terminal sends the input instruction to the server.

[1260] Input: Instruction text data.

[1261] Output: The HTTP POST request sent to the server.

[1262] Specific behavior: The input text is included in the body of the HTTP request and sent to the specified URL.

[1263] Step 3:

[1264] The server receives the instructions and passes them to the generative artificial intelligence.

[1265] Input: The HTTP POST request sent from the device.

[1266] Output: API request to the generative AI.

[1267] Specific operation: The server extracts instructions from the body of the HTTP request and sends the request to the API of the generative artificial intelligence (e.g., OpenAI GPT series).

[1268] Step 4:

[1269] Generative artificial intelligence generates answers based on instructions.

[1270] Input: Instruction text data.

[1271] Output: The generated answer text.

[1272] How it works: Generative AI uses natural language processing to generate appropriate answers based on the user's instructions. For example, it generates an answer such as, "In order to adapt to a new environment, it is important to actively communicate with those around you and strengthen self-management."

[1273] Step 5:

[1274] The server receives the generated answer and sends it to the ethical check artificial intelligence.

[1275] Input: Answer text received from the generative AI.

[1276] Output: API request to the Ethics Check AI.

[1277] Specific operation: The server extracts the generated answers and sends a request to the API of the ethical check artificial intelligence (e.g., filtering model).

[1278] Step 6:

[1279] Ethics check: Artificial intelligence evaluates the appropriateness of generated answers.

[1280] Input: Answer text data.

[1281] Output: The result of evaluating whether the answer is correct.

[1282] What it does: Ethics check AI analyzes your answers and checks for inappropriate content or misinformation.

[1283] Step 7:

[1284] The server receives the evaluation results of the ethical check artificial intelligence and, if necessary, sends a regeneration request to the generative artificial intelligence.

[1285] Input: Evaluation results from Ethics Check Artificial Intelligence.

[1286] Output: Repeat request to generative AI if necessary.

[1287] Specific operation: If the answer is determined to be inappropriate, the server sends a regeneration request to the generative artificial intelligence to generate a new answer.

[1288] Step 8:

[1289] The server formats the accepted response in the output processing module.

[1290] Input: The accepted answer text.

[1291] Output: The formatted answer text.

[1292] Specific behavior: Converts the accepted answer text into a user-readable format (e.g., HTML format).

[1293] Step 9:

[1294] The server sends the formatted response to the user's terminal.

[1295] Input: The formatted answer text.

[1296] Output: The HTTP response sent to the user's device.

[1297] Specific behavior: Sends formatted text to the terminal as an HTTP response.

[1298] Step 10:

[1299] The terminal displays the received answer to the user.

[1300] Input: The HTTP response received from the server.

[1301] Output: The answer text that is displayed on the screen.

[1302] Specific operation: Display the received data in an easy-to-read format in a web browser or application interface.

[1303] Through these steps, users can receive safe and relevant information quickly.

[1304] (Application example 1)

[1305] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1306] In factories, workers often have questions or problems regarding production lines, machine operation, and quality control, and it is difficult to obtain appropriate answers each time. This reduces work efficiency and increases the risk of operational errors. Furthermore, utilizing human expertise requires a great deal of time and effort.

[1307] 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.

[1308] In this invention, the server includes means for receiving prompts input by a user, generative AI means for generating answers based on the prompts, morality check AI means for evaluating the appropriateness of the generated answers, means for transmitting answers approved by the morality check AI means to an output device, and means for providing answers to questions from factory workers through the output device, thereby making it possible to provide factory workers with prompt and appropriate information in response to their doubts and problems.

[1309] A "user" is a person who uses a system or device and is a worker who is responsible for inputting prompts.

[1310] A "prompt" refers to a question or instruction entered by a user, and is input information used by generative artificial intelligence to generate an answer.

[1311] "Generative AI" is an AI technology that generates appropriate answers based on prompts entered by the user.

[1312] "Moral check AI" is an AI technology that evaluates whether the generated answers are ethically appropriate.

[1313] "Output device" refers to a device for providing the user with the final answer generated by the generative artificial intelligence and the moral check artificial intelligence.

[1314] "Factory worker" means a person who operates production lines and machines, performs quality control, and other tasks within a factory.

[1315] A "question" is a question or inquiry for information that a factory worker has, and is entered as a prompt.

[1316] An "answer" is information generated by the generative AI based on a prompt and provided after being evaluated by the moral check AI.

[1317] This invention is a system that provides prompt and appropriate information to factory workers in response to questions and problems. This system utilizes generative AI and moral check AI to efficiently generate answers and evaluate their appropriateness.

[1318] The system is structured as follows: First, a user inputs a question for a factory worker as a prompt. In response, the input prompt is sent to the server. The server uses generative AI to generate an answer to the prompt. The generated answer is then evaluated for appropriateness by morality check AI.

[1319] The answers that have been approved are sent from the server to an output device and provided to the factory workers. If they are deemed inappropriate, they are regenerated or corrected. For example, OpenAI's API is used for generative AI. Dedicated natural language processing technology is used for moral check AI.

[1320] The specific system operation procedure is as follows: A factory worker speaks to the robot, saying, "Please tell me how to operate this machine." The robot receives this prompt and sends it to the server. The generative AI generates an answer such as, "To operate this machine, first turn it on, then follow the setup guide and press the buttons." The appropriateness of the answer is evaluated by the moral check AI, and if it is deemed appropriate, the robot provides the answer to the factory worker.

[1321] This system allows factory workers to efficiently obtain the appropriate information. Below are some example prompts:

[1322] Example prompt:

[1323] 1. "How do I set up a new production line?"

[1324] 2. "Where are the quality checkpoints for this product?"

[1325] 3. "Please explain how to deal with mechanical problems."

[1326] This will improve work efficiency within the factory and reduce the risk of operational errors and problems.

[1327] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1328] Step 1: The user enters the prompt.

[1329] Input: A factory worker speaks a question to the robot.

[1330] Output: The robot receives the user's prompt and converts it into a data format.

[1331] Specific operation: When a factory worker speaks to the robot, saying, "Please tell me how to operate this machine," the robot's voice recognition function converts the question into text data.

[1332] Step 2: Send the entered prompt from the terminal to the server.

[1333] Input: The prompt converted into text data by the robot.

[1334] Output: Text data is sent to the server.

[1335] Specific operation: The robot converts the prompt text data into an HTTP request format and sends it to the server.

[1336] Step 3: The server generates an answer using generative artificial intelligence.

[1337] Input: The prompt data sent to the server.

[1338] Output: Answer text generated by generative artificial intelligence.

[1339] Specific operation: The server passes the received prompt data to the OpenAI API, and the generative AI model generates an answer such as, "To operate this machine, first turn it on, then press the buttons according to the setup guide."

[1340] Step 4: The generated answers are sent to the moral check AI for evaluation.

[1341] Input: Answer text generated by generative artificial intelligence.

[1342] Output: Evaluation result (approval or disapproval) by the moral check AI.

[1343] Specific operation: The server sends the generated answer to the moral check AI, which uses natural language processing technology to evaluate the appropriateness of the answer and check whether it contains inappropriate content or misinformation.

[1344] Step 5: Revise or regenerate answers based on the results of the moral check AI.

[1345] Input: The generated answer when the evaluation result of the moral check artificial intelligence is disapproval.

[1346] Output: The corrected or regenerated answer text.

[1347] Specific operation: If the moral check AI judges the generated answer to be inappropriate, the server will again prompt the generator AI to generate a new answer. It may also request a regeneration by adding a phrase such as "Please try again."

[1348] Step 6: Moral Check The answer approved by the AI ​​is sent to the terminal.

[1349] Input: Answer text approved by the Morality Check AI.

[1350] Output: The accepted answer is sent to the robot.

[1351] Specific operation: The server sends the final accepted answer to the robot as an HTTP response.

[1352] Step 7: The robot provides the answer to the factory worker.

[1353] Input: The accepted answer text received from the server.

[1354] Output: Provides information to factory workers.

[1355] Specific operation: The robot uses a voice synthesis function to audibly convey the received response to the factory worker. For example, it might say something like, "To operate this machine, first turn it on, then follow the setup guide and press the buttons."

[1356] Through the above steps, factory workers can efficiently obtain appropriate information, thereby improving the accuracy and efficiency of their work.

[1357] 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.

[1358] This invention uses generative artificial intelligence (AI) to generate answers to prompts entered by users, evaluates the appropriateness of those answers, and combines this with an emotion engine that recognizes the user's emotions to provide more accurate and appropriate information. This system allows users to receive safe and appropriate information in a style appropriate to their emotions.

[1359] System Configuration

[1360] The system consists of the following main components:

[1361] 1. Input Processing Module

[1362] 2. Generative AI

[1363] 3. Moral Check AI

[1364] 4. Emotion Engine

[1365] 5. Output Processing Module

[1366] User prompt input

[1367] The user uses the terminal to enter a prompt, for example, "Tell me how to adapt to a new environment," and this prompt is sent from the terminal to the server.

[1368] Server-sent prompts

[1369] The terminal sends the entered prompt to the server as an HTTP request.

[1370] Emotion Analysis

[1371] The server passes the received prompt to the emotion engine, which analyzes the user's emotion. For example, it analyzes that the user is feeling "anxiety" based on the context and keywords of the text.

[1372] Answer generation using generative AI

[1373] The server passes the prompt, including the analyzed emotional information, to the generative AI. The generative AI generates an answer based on the prompt and the emotional information. For example, it generates an answer such as, "To adapt to a new environment, it is important to actively communicate with those around you and strengthen self-management. Don't worry, just take your time to get used to it."

[1374] Server reception of response

[1375] The server receives the generated answer and then sends the answer to the morale check AI.

[1376] Evaluation by moral check AI

[1377] The server sends the generated answer to the morality check AI, which evaluates the appropriateness of the answer. The morality check AI uses natural language processing technology to check whether the answer contains inappropriate content or misinformation. For example, the morality check AI evaluates the answer and determines that it is appropriate.

[1378] Corrective action (if necessary)

[1379] The server receives the evaluation results from the moral check AI and sends the approved answer directly to the user. On the other hand, if the answer is judged to be inappropriate, it sends a request to the generative AI again based on the moral check AI's suggested corrections, and a new answer is generated.

[1380] Server processing of final response

[1381] The server receives the final approved response and formats it in an appropriate display format in the output processing module.

[1382] User-submitted answers

[1383] The server sends the final answer to the user's terminal as an HTTP response.

[1384] Show Answers

[1385] The terminal displays the received answer to the user in an easy-to-read format on the terminal interface (web browser, application, etc.).

[1386] Specific examples

[1387] For example:

[1388] The user types "Tell me how to adapt to a new environment" into their device. The device sends this prompt to the server. The server passes the prompt to the emotion engine, which analyzes the user's emotion as "anxiety." The server passes the prompt along with the analyzed emotion information to the generative AI. The generative AI generates an answer: "To adapt to a new environment, it's important to actively communicate with those around you and strengthen self-management. Don't worry, just get used to it slowly." The server passes the generated answer to the morality check AI, which evaluates the answer and determines it is appropriate. The server sends the approved answer to the user's device, and the device displays the answer to the user.

[1389] This system allows users to quickly and safely receive appropriate information corresponding to their emotions, without being aware of the process that the generative AI is generating.

[1390] The processing flow will be explained below.

[1391] Step 1:

[1392] The user types a prompt into the terminal, for example, "Tell me how to adapt to a new environment."

[1393] Step 2:

[1394] The terminal sends the entered prompt to the server as an HTTP request.

[1395] Step 3:

[1396] The server passes the received prompt to the emotion engine, which analyzes the user's emotion. For example, it analyzes that the user is feeling "anxiety" based on the context and keywords of the text.

[1397] Step 4:

[1398] The server passes the prompt containing the analyzed emotional information to the generative AI. For example, the emotional information "anxiety" is sent along with the prompt.

[1399] Step 5:

[1400] Generative AI generates answers that take emotional information into account, such as, "To adapt to a new environment, it's important to actively communicate with those around you and strengthen self-management. Don't worry, just take your time to get used to it."

[1401] Step 6:

[1402] The server receives the answer generated by the generative AI.

[1403] Step 7:

[1404] The server sends the generated answer to the morality check AI, for example, by sending a request to the morality check AI's API to check whether the generated answer is appropriate.

[1405] Step 8:

[1406] The moral check AI analyzes the answers it receives using natural language processing technology and evaluates the appropriateness of the answers, for example, checking to see if they contain inappropriate language or misinformation.

[1407] Step 9:

[1408] The server receives the evaluation results from the moral check AI. Approved answers proceed to the next step, while answers deemed inappropriate are sent to the generative AI for revision, which then generates a new answer.

[1409] Step 10:

[1410] The server then formats the accepted response in an output processing module into an appropriate display format, for example, HTML or JSON.

[1411] Step 11:

[1412] The server sends the final answer to the user's device as an HTTP response.

[1413] Step 12:

[1414] The device will display the received answer to the user. For example, the answer will be displayed in an easy-to-read format in a web browser or application. Specifically, the device will display a message such as, "To adapt to a new environment, it is important to actively communicate with those around you and strengthen self-management. Don't worry, just take your time to get used to it."

[1415] Example 2

[1416] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1417] Conventional generative AI systems generate answers that do not sufficiently consider the user's emotions, which can result in the system failing to provide the appropriate information the user is looking for. Furthermore, there is a lack of a process for determining the appropriateness of the generated answers, which creates a risk of providing inappropriate information to the user. As a result, there is a risk of the system losing user trust.

[1418] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes means for receiving a prompt input by a user, emotion engine means for analyzing the prompt and recognizing the user's emotion, generative AI means for generating an answer based on the prompt including the analyzed emotion information, morality check AI means for evaluating the appropriateness of the generated answer, means for transmitting the answer approved by the morality check AI means to a terminal, and means for displaying the answer on the terminal. This makes it possible to provide appropriate and safe information according to the user's emotion.

[1419] A "prompt" is text data such as a question or instruction entered by a user.

[1420] An "emotion engine" is software or hardware that analyzes a user's prompts and recognizes the user's emotions from their context and keywords.

[1421] "Generative AI" is an AI model for generating answers in natural language based on prompts and emotional information.

[1422] "Moral check AI" is an AI model that evaluates the appropriateness of answers generated by generative AI and checks whether they contain inappropriate content or misinformation.

[1423] A "terminal" is a device (such as a PC or smartphone) on which a user enters prompts and receives generated answers.

[1424] This invention combines a system that uses generative artificial intelligence (AI) to generate answers to prompts entered by users and evaluates whether the answers are appropriate, with an emotion engine that recognizes the user's emotions. This system allows users to receive safe and appropriate information in a style appropriate to their emotions.

[1425] System Configuration

[1426] The system consists of the following main components:

[1427] 1. Input Processing Module

[1428] 2. Generative AI

[1429] 3. Moral Check AI

[1430] 4. Emotion Engine

[1431] 5. Output Processing Module

[1432] User prompt input

[1433] The user uses the terminal to enter a prompt. Specifically, the user types, "Tell me how to adapt to a new environment." This prompt is sent from the terminal to the server.

[1434] Server-sent prompts

[1435] The terminal sends the entered prompt to the server as an HTTP request.

[1436] Emotion Analysis

[1437] The server passes the received prompt to the emotion engine, which analyzes the user's emotion. The emotion engine, for example, uses a natural language processing toolkit to analyze the context and keywords in the text to determine that the user is feeling "anxiety." This information is converted into JSON format and passed on to the next process.

[1438] Answer generation using generative AI

[1439] The server passes the prompt, including the analyzed emotional information, to the generative AI. Specifically, new JSON data with the emotion engine results added is sent as input to the generative AI. The generative AI then generates an answer based on the prompt and emotional information, using, for example, OpenAI's GPT model. An example of a generated answer would be, "To adapt to a new environment, it's important to actively communicate with those around you and strengthen self-management. Don't worry, just take it easy and get used to it."

[1440] Server reception of response

[1441] The server receives the generated answer and then sends it to the moral check AI. The response from the generating AI comes back in JSON format, which the server parses.

[1442] Evaluation by moral check AI

[1443] The server sends the generated answers to a morality check AI, which evaluates their appropriateness. The morality check AI analyzes the answers created by the generative AI, for example, using a model with specific ethical filters, to check for inappropriate content or misinformation. If the morality check AI determines that the answers are appropriate, this information is also returned in JSON format.

[1444] Corrective action (if necessary)

[1445] The server receives the evaluation results from the morality check AI. If the answer is judged to be inappropriate, it sends a request to the generative AI again based on the correction suggestions provided by the morality check AI to generate a new answer. Specifically, it appropriately combines the original prompt with the morality check AI's feedback and sends a new JSON request to the generative AI.

[1446] Server processing of final response

[1447] The server receives the final approved answer and formats it in an appropriate display format using an output processing module, specifically, formatting the answer in a data format such as HTML or JSON, converting it into a format that is easy for users to understand.

[1448] User-submitted answers

[1449] The server sends the final answer to the user's device as an HTTP response. Specifically, it returns the formatted data to the device again via the HTTPS protocol.

[1450] Show Answers

[1451] The device displays the received response to the user. Specifically, the web browser or application analyzes the received data and displays it in the appropriate location on the screen. The user can see the following text: "To adapt to a new environment, it is important to actively communicate with those around you and strengthen self-management. Don't worry, just take it easy and get used to it."

[1452] This system allows users to quickly and safely receive appropriate information corresponding to their emotions, without being aware of the process that the generative AI is generating.

[1453] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1454] Step 1:

[1455] The user inputs the prompt using their own device. Specifically, the user types "Please tell me how to adapt to the new environment" into the input field of the device's web browser or dedicated application, and clicks the send button. The input data is the prompt text in text format.

[1456] Step 2:

[1457] The terminal sends the entered prompt to the server as an HTTP request. Specifically, JSON format data containing the user input is sent via the HTTPS protocol. The input data is the user prompt, and the output data is the HTTP request sent to the server.

[1458] Step 3:

[1459] The server passes the received prompt to the emotion engine and analyzes the user's emotion. Specifically, the emotion engine (which includes a natural language processing toolkit) analyzes the text and determines from keywords and context that the user is feeling "anxiety." The input data is the received prompt, and the output data is JSON-formatted data containing the analyzed emotion information.

[1460] Step 4:

[1461] The server passes the prompt containing the analyzed emotional information to the generative AI. Specifically, new JSON data with the results of the emotion engine added is sent as input to the generative AI. The generative AI generates an answer based on this. The input data is the prompt containing emotional information, and the output data is JSON format data containing the generated answer.

[1462] Step 5:

[1463] The server receives the generated answer and then sends it to the morality check AI. The response from the generation AI comes back in JSON format, which the server parses. The input data is the generated answer, and the output data is the data sent to the morality check AI.

[1464] Step 6:

[1465] The server sends the generated answer to the morality check AI, which evaluates the appropriateness of the answer. Specifically, the morality check AI (a model with specific ethical filters) analyzes the answer and checks whether it contains inappropriate content or misinformation. If the morality check AI determines that the answer is appropriate, this information is also returned in JSON format. The input data is the generated answer, and the output data is data containing the evaluation result of the appropriateness.

[1466] Step 7:

[1467] The server receives the evaluation results from the morality check AI. If the answer is judged to be inappropriate, it sends a request to the generative AI again based on the correction suggestions provided by the morality check AI to generate a new answer. Specifically, it appropriately combines the original prompt and the morality check AI's feedback and sends a new JSON request to the generative AI. The input data is the evaluation result from the morality check AI, and the output data is the corrected prompt.

[1468] Step 8:

[1469] The server finally receives the approved answer and formats it in the appropriate display format in the output processing module. Specifically, the answer content is formatted in data formats such as HTML or JSON and converted into a form that is easy for users to understand. The input data is the approved answer, and the output data is the formatted answer.

[1470] Step 9:

[1471] The server sends the final answer to the user's terminal as an HTTP response. Specifically, it returns the formatted data to the terminal again via the HTTPS protocol. The input data is the formatted answer, and the output data is the HTTP response.

[1472] Step 10:

[1473] The device displays the received response to the user. Specifically, the web browser or application analyzes the received data and displays it in the appropriate location on the screen. The user can see the following text: "To adapt to a new environment, it is important to actively communicate with those around you and strengthen self-management. Don't worry, just take it easy and get used to it." The input data is the HTTP response, and the output data is the text displayed to the user.

[1474] (Application example 2)

[1475] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1476] Conventional food delivery systems simply present a menu without considering the user's emotions or mood when ordering. This results in a lack of flexible suggestions and advice based on the user's current mental state and emotions, leading to a lack of satisfaction in the user experience. Furthermore, the inability to provide appropriate information based on the user's emotions makes it difficult to provide the comfortable service that users desire.

[1477] 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.

[1478] In this invention, the server includes means for receiving a prompt input by a user, generative AI means for generating an answer based on the prompt, emotion analysis means for providing emotion analysis information to the generative AI means, morality check AI means for evaluating the appropriateness of the generated answer, and means for transmitting to a terminal recommendation information based on the answer approved by the morality check AI means and the emotion analysis information, thereby making it possible to provide the user with appropriate meal menu suggestions and advice in the form of customized messages according to their emotions and moods.

[1479] "User" refers to an individual or organization that uses this system.

[1480] A "prompt" is a request or question that a user enters into a system.

[1481] "Generative artificial intelligence means" refers to AI technologies that generate appropriate answers based on user prompts.

[1482] "Emotion analysis means" refers to a technique for analyzing emotion information from a prompt entered by a user.

[1483] "Moral check artificial intelligence means" refers to AI technology that evaluates the appropriateness of generated answers and eliminates inappropriate content.

[1484] "Terminal" refers to a device for providing generated answers and recommendation information to a user.

[1485] "Emotion analysis information" refers to data that indicates the emotional state of the user extracted by the emotion analysis means.

[1486] "Recommended information" refers to specific options or advice suggested to users based on sentiment analysis information.

[1487] The present invention relates to a system that generates answers to prompts entered by a user, evaluates the appropriateness of the answers, and provides appropriate information based on the user's emotions. In particular, the present invention is applied to food delivery, and provides appropriate menu suggestions and emotional support messages by taking into account the user's emotions when ordering a meal.

[1488] System configuration

[1489] The system consists of the following main components:

[1490] 1. Input Processing Module: A user uses a terminal to input a prompt, for example, the user inputs "I'm feeling down today."

[1491] 2. Generative AI methods: Generate appropriate answers based on user prompts. For example, "If you're looking for something to cheer you up, I recommend chicken soup or salad."

[1492] 3. Emotion analysis: Analyzes emotional information from user input. Example: Analyzes "I feel depressed."

[1493] 4. Moral Check AI Method: Evaluate the appropriateness of the generated answer. For example, check whether the generated answer is harmful to the user’s feelings.

[1494] 5. Output processing module: Sends the final answer and recommendation information to the terminal and displays it. Example: Display "If you're looking for something to cheer you up, we recommend chicken soup or salad."

[1495] Hardware and software used

[1496] Hardware: User devices (smartphones, tablets, etc.), servers

[1497] Software: Python, requests library, emotion analysis module, generative AI module, moral check module, menu recommendation module

[1498] Data processing and calculation

[1499] 1. User prompt input: The terminal receives the user prompt and sends it to the server as an HTTP request.

[1500] 2. Sentiment analysis: The server passes the prompt received to the sentiment analyzer, which analyzes the user's sentiment based on the context and keywords in the text. For example, "I feel depressed" is analyzed and determined to be "negative."

[1501] 3. Answer generation by generative AI: The prompt containing the analyzed emotional information is passed to a generative AI means to generate an answer. The generated answer is in line with the user's emotions. Example: "If you're looking for something to cheer you up, I recommend chicken soup or salad."

[1502] 4. Moral check: The generated answer is sent to the moral check AI means to evaluate its appropriateness. If it is judged to be inappropriate, a correction suggestion is regenerated. Example: The generated answer "Chicken soup or salad is recommended" is judged to be appropriate.

[1503] 5. Sending the final answer: The output processing module formats the accepted answer and sends it to the user's device as an HTTP response. The user's device displays the text "If you're looking for something to cheer you up, I recommend chicken soup or salad."

[1504] Specific examples

[1505] A user uses a food delivery assistant application and inputs, "I'm feeling down today." The device sends this to the server. The server receives the prompt, and the emotion analysis means analyzes the emotional information (negative). Based on the analyzed emotional information, the generative AI generates an answer: "If you're looking for something to cheer you up, I recommend chicken soup or salad." After the moral check AI determines this answer is appropriate, the final answer is sent to the user's device and displayed.

[1506] Example of a user prompt:

[1507] "I'm feeling down today, so I'd like you to recommend a dish that will cheer me up."

[1508] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1509] Step 1:

[1510] The user inputs a prompt through the terminal. Specifically, the user inputs "I'm feeling depressed today." This input text is acquired as prompt information.

[1511] Input: User prompt (e.g., "I'm feeling down today")

[1512] Output: Prompt information (text)

[1513] Step 2:

[1514] The terminal sends the input prompt information to the server as an HTTP request. Specifically, the prompt information is sent to the server's input processing module using a REST API or similar.

[1515] Input: Prompt information (text)

[1516] Output: Prompt information sent to the server (HTTP request)

[1517] Step 3:

[1518] The server passes the received prompt information to the emotion analysis means, which analyzes the user's emotion. Specifically, it uses a text analysis algorithm to recognize the user's emotion (e.g., "negative") from the prompt content.

[1519] Input: Prompt information (text)

[1520] Output: Sentiment analysis information (e.g., "negative")

[1521] Step 4:

[1522] The server passes prompt information, including the analyzed emotional information, to the generative AI means. The generative AI model generates an appropriate response based on the prompt information and emotional information. Specifically, in response to the statement "I'm feeling down today," the AI ​​model generates the response "If you're looking for something to cheer you up, I recommend chicken soup or a salad."

[1523] Input: prompt information (text), sentiment analysis information (e.g., "negative")

[1524] Output: Generated answer (e.g., "For comfort food, chicken soup or salad is recommended.")

[1525] Step 5:

[1526] The server sends the generated answer to the morality check AI means, which evaluates the appropriateness of the answer. Specifically, the morality check AI checks whether the generated answer contains inappropriate content or misinformation, and returns an evaluation result of whether the answer is appropriate.

[1527] Input: Generated answer (text)

[1528] Output: Evaluation result (appropriate / inappropriate)

[1529] Step 6:

[1530] The server receives the evaluation results from the moral check AI, and if the evaluation results are appropriate, it sends the final answer and recommendation information to the device. Specifically, it sends the answer that it judges to be appropriate (e.g., "If you're looking for something to cheer you up, we recommend chicken soup or salad.") to the user's device as an HTTP response.

[1531] Input: Evaluation results, generated answers (text)

[1532] Output: Final answer (text) sent to the terminal

[1533] Step 7:

[1534] The device displays the final answer it received to the user. Specifically, it displays "If you're looking for something to cheer you up, we recommend chicken soup or salad" in an easy-to-read format on the device's UI (user interface).

[1535] Input: Final answer (text)

[1536] Output: what is displayed to the user (text)

[1537] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1538] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1539] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1540] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1541] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1542] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1543] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1544] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1545] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1546] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1547] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1548] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1549] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1550] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[1551] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1552] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1553] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1554] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1555] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1556] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1557] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1558] The following is further disclosed regarding the above embodiment.

[1559] (Claim 1)

[1560] means for receiving a user-entered prompt;

[1561] a generative artificial intelligence means for generating answers based on the prompts;

[1562] a morality check artificial intelligence means for evaluating the appropriateness of the generated answers;

[1563] means for transmitting the answer approved by the morality check artificial intelligence means to a terminal;

[1564] A system including:

[1565] (Claim 2)

[1566] 2. The system according to claim 1, further comprising means for returning a request for correction or regeneration to said generative artificial intelligence means when the generated answer is determined to be inappropriate.

[1567] (Claim 3)

[1568] 10. The system of claim 1, further comprising: means for converting the prompts and generated answers into an appropriate format; and means for formatting the answers before displaying them on the terminal.

[1569] "Example 1"

[1570] (Claim 1)

[1571] means for receiving instructions input by a user;

[1572] means for transmitting the instruction to a server;

[1573] A generative artificial intelligence means for generating an answer based on the instruction sentence;

[1574] an ethics check artificial intelligence means for evaluating the appropriateness of the generated answers;

[1575] means for transmitting the answer approved by the ethical check artificial intelligence means to a terminal;

[1576] means for the terminal to display the answer to the user;

[1577] A system including:

[1578] (Claim 2)

[1579] 2. The system according to claim 1, further comprising means for returning a request for correction or regeneration to said generative artificial intelligence means when the generated answer is determined to be inappropriate.

[1580] (Claim 3)

[1581] 10. The system of claim 1, further comprising: means for converting the instructions and generated answers into an appropriate format; and means for formatting the answers before displaying them on the terminal.

[1582] "Application Example 1"

[1583] (Claim 1)

[1584] means for receiving a user-entered prompt;

[1585] a generative artificial intelligence means for generating answers based on the prompts;

[1586] a morality check artificial intelligence means for evaluating the appropriateness of the generated answers;

[1587] means for transmitting the answer approved by the morality check artificial intelligence means to an output device;

[1588] means for providing answers to questions from factory workers through said output device;

[1589] A system including:

[1590] (Claim 2)

[1591] 2. The system according to claim 1, further comprising means for returning a request for correction or regeneration to said generative artificial intelligence means when the generated answer is determined to be inappropriate.

[1592] (Claim 3)

[1593] 10. The system of claim 1, further comprising: means for converting the prompts and generated answers into an appropriate format; and means for formatting the answers before displaying them on the output device.

[1594] "Example 2: Combining Emotion Engines"

[1595] (Claim 1)

[1596] means for receiving a user-entered prompt;

[1597] emotion engine means for analyzing the prompt and recognizing the emotion of the user;

[1598] a generative artificial intelligence means for generating an answer based on the prompt including the analyzed emotion information;

[1599] a morality check artificial intelligence means for evaluating the appropriateness of the generated answers;

[1600] means for transmitting the answer approved by the morality check artificial intelligence means to a terminal;

[1601] means for displaying an answer on the terminal;

[1602] A system including:

[1603] (Claim 2)

[1604] 2. The system according to claim 1, further comprising means for returning a request for correction or regeneration to said generative artificial intelligence means when the generated answer is determined to be inappropriate.

[1605] (Claim 3)

[1606] 10. The system of claim 1, further comprising: means for converting the prompts and generated answers into an appropriate format; and means for formatting the answers before displaying them on the terminal.

[1607] "Application example 2 when combining emotion engines"

[1608] (Claim 1)

[1609] means for receiving a user-entered prompt;

[1610] a generative artificial intelligence means for generating answers based on the prompts;

[1611] emotion analysis means for providing emotion analysis information to the generative artificial intelligence means;

[1612] a morality check artificial intelligence means for evaluating the appropriateness of the generated answers;

[1613] means for transmitting to a terminal recommendation information based on the answer approved by the moral check artificial intelligence means and emotion analysis information;

[1614] A system including:

[1615] (Claim 2)

[1616] 2. The system according to claim 1, further comprising means for returning a request for correction or regeneration to said generative artificial intelligence means when the generated answer is determined to be inappropriate.

[1617] (Claim 3)

[1618] 10. The system of claim 1, further comprising: means for converting the prompts and generated answers into an appropriate format; means for formatting the answers before displaying them on the terminal; and means for providing recommendations based on sentiment analysis information. [Explanation of symbols]

[1619] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. means for receiving a user-entered prompt; a generative artificial intelligence means for generating answers based on the prompts; a morality check artificial intelligence means for evaluating the appropriateness of the generated answers; means for transmitting the answer approved by the morality check artificial intelligence means to a terminal; A system including:

2. 2. The system according to claim 1, further comprising means for returning a request for correction or regeneration to said generative artificial intelligence means when the generated answer is determined to be inappropriate.

3. 2. The system of claim 1, further comprising: means for converting the prompts and generated answers into an appropriate format; and means for formatting the answers before displaying them on the terminal.

Images