system

Abstract

The system according to this embodiment aims to efficiently operate a certification system for AI agents. [Solution] The system according to the embodiment comprises an application acceptance unit, a testing unit, an examination unit, a certification unit, and an audit unit. The application acceptance unit accepts applications. The testing unit tests the applications accepted by the application acceptance unit. The examination unit examines the results of the tests conducted by the testing unit. The certification unit makes certifications based on the results of the examination conducted by the examination unit. The audit unit audits the activities of the AI ​​agents certified by the certification unit.

Description

Technical Field

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

Background Art

[0002] Patent Document 1 discloses a persona chatbot control method performed by at least one processor, including steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to an explanation of a 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

Summary of the Invention

Problems to be Solved by the Invention

[0004] In the conventional technology, there is a problem that it is limited in terms of resources and capabilities for humans to operate the certification qualification system of AI agents.

[0005] The system according to the embodiment aims to efficiently operate the certification qualification system of AI agents.

Means for Solving the Problems

[0006] The system according to this embodiment comprises an application acceptance unit, a testing unit, an examination unit, an accreditation unit, and an audit unit. The application acceptance unit accepts applications. The testing unit tests the applications accepted by the application acceptance unit. The examination unit reviews the results of the tests conducted by the testing unit. The accreditation unit accredits the applications based on the results of the examination conducted by the examination unit. The audit unit audits the activities of the AI ​​agents accredited by the accreditation unit. [Effects of the Invention]

[0007] The system according to this embodiment can efficiently operate a certification system for AI agents. [Brief explanation of the drawing]

[0008] [Figure 1] This is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] This is a conceptual diagram showing an example of the essential functions of a data processing device and a smart device according to the first embodiment. [Figure 3] This is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] This is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment. [Figure 5] This is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] This is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7] This is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] This is a conceptual diagram showing an example of the main functions of a data processing device and a robot according to the fourth embodiment. [Figure 9] This shows an emotion map where multiple emotions are mapped. [Figure 10] This shows an emotion map where multiple emotions are mapped. [Modes for carrying out the invention]

[0009] Hereinafter, an example of an embodiment of the system relating to the technology of this disclosure will be described with reference to the attached drawings.

[0010] First, let's explain the terminology used in the following explanation.

[0011] In the following embodiments, the signed processor (hereinafter simply referred to as "processor") may be a single arithmetic unit or a combination of multiple arithmetic units. Furthermore, the processor may be a single type of arithmetic unit or a combination of multiple types of arithmetic units. Examples of arithmetic units include CPU (Central Processing Unit), GPU (Graphics Processing Unit), GPGPU (General-Purpose computing on Graphics Processing Units), APU (Accelerated Processing Unit), or TPU (Tensor Processing Unit).

[0012] In the following embodiments, signed RAM (Random Access Memory) is a memory that temporarily stores information and is used as work memory by the processor.

[0013] In the following embodiments, the signed storage is one or more non-volatile storage devices that store various programs and various parameters. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes.

[0014] In the following embodiments, the numbered communication I / F (Interface) is an interface including a communication processor, an antenna, etc. The communication I / F manages communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards such as 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or Bluetooth (registered trademark).

[0015] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B". That is, "A and / or B" means that it may be only A, only B, or a combination of A and B. Also, in this specification, when expressing three or more matters connected by "and / or", the same concept as "A and / or B" is applied.

[0016] [First Embodiment] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0017] As shown in FIG. 1, the 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.

[0018] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. 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. Also, the database 24 and the communication I / F 26 are 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).

[0019] The smart device 14 comprises a computer 36, a receiving device 38, an output device 40, a camera 42, and a communication interface 44. The computer 36 comprises a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The receiving device 38, output device 40, and camera 42 are also connected to the bus 52.

[0020] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, and accepts user input. The touch panel 38A accepts user input via touch by detecting contact with an object (e.g., a pen or finger). The microphone 38B accepts user input via voice by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and microphone 38B to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 (see Figure 2) acquires the data indicating the user input.

[0021] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user by outputting the data in a form perceptible to the user (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0022] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

[0023] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0024] As shown in Figure 2, in the data processing device 12, a specific processing 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" related to the technology of this 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 according to the specific processing program 56 executed on the RAM 30.

[0025] Storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290. The identification processing unit 290 can estimate the user's emotions using the emotion identification model 59 and perform identification processing using the user's emotions. The emotion estimation function (emotion identification function) using the emotion identification model 59 performs various estimations and predictions regarding the user's emotions, including but not limited to these examples. Furthermore, emotion estimation and prediction also include, for example, emotion analysis.

[0026] In the smart device 14, specific processing is performed by the processor 46. The storage 50 stores a specific processing program 60. The specific processing program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the specific processing program 60 from the storage 50 and executes the read specific processing program 60 on the RAM 48. The specific processing is realized by the processor 46 operating as a control unit 46A according to the specific processing program 60 executed on the RAM 48. The smart device 14 also has a data generation model 58 and an emotion identification model 59, similar to the data generation model and emotion identification model 59, and can perform processing similar to that of the specific processing unit 290 using these models.

[0027] Furthermore, other devices besides the data processing device 12 may also have the data generation model 58. For example, a server device (e.g., a generation server) may have the data generation model 58. In this case, the data processing device 12 obtains processing results (such as prediction results) using the data generation model 58 by communicating with the server device having the data generation model 58. The data processing device 12 may also be a server device or a terminal device owned by a user (e.g., a mobile phone, robot, home appliance, etc.). Next, an example of processing by the data processing system 10 according to the first embodiment will be described.

[0028] (Example of form 1) An AI agent certification system according to an embodiment of the present invention organizes a group of AI agents for operating an AI agent certification system and provides an agency that leads the agents responsible for each task. In the AI ​​agent certification system, the application acceptance department accepts certification applications for AI agents, the examination department tests the skills and ethical aspects of the applied AI agents, the review department reviews the test results, the certification department grants certification, and the audit department audits the activities of the certified AI agents. Furthermore, to verify the behavioral and conversational capabilities of the AI ​​agents, collaboration with a physical robot is introduced. For example, the AI ​​agent gives instructions to the physical robot, and the robot performs actions based on those instructions. The autonomous examiner AI agent evaluates the robot's actions and verifies their accuracy and practicality. This ensures that the AI ​​agent's instructions are accurate and practical. Furthermore, the autonomous examiner AI agent autonomously performs evaluation tasks and provides feedback. For example, it uses generative AI to engage in dialogue and evaluates the AI ​​agent's conversational ability during that dialogue. In addition, the generative AI analyzes the AI ​​agent's instructions and gives appropriate action instructions to the physical robot, thereby evaluating the accuracy of instruction understanding and execution. This mechanism allows for the efficient operation of the AI ​​agent certification system and improves the practicality and reliability of AI agents. For example, by using physical robots to evaluate the accuracy and practicality of AI agent execution and providing accurate and efficient feedback, the practicality and reliability of AI agents are enhanced. As a result, the AI ​​agent certification system can efficiently operate the AI ​​agent certification system and improve the practicality and reliability of AI agents.

[0029] The AI ​​agent certification system according to this embodiment comprises an application acceptance unit, an examination unit, an evaluation unit, a certification unit, and an audit unit. The application acceptance unit accepts applications. The application acceptance unit can accept applications in various forms, such as paper applications and online applications. The examination unit tests the applications accepted by the application acceptance unit. The examination unit tests, for example, the skills and ethical aspects of the AI ​​agent. The examination unit can conduct tests to evaluate the programming skills and problem-solving abilities of the AI ​​agent. The examination unit can also conduct tests to evaluate the ethical judgment abilities of the AI ​​agent. The evaluation unit evaluates the results of the tests conducted by the examination unit. The evaluation unit evaluates, for example, the accuracy and reliability of the test results. The evaluation unit determines whether or not to certify the AI ​​agent based on the test results. The certification unit certifies the AI ​​agent based on the results of the evaluation conducted by the evaluation unit. The certification unit certifies the AI ​​agent, for example, by confirming whether the skills and ethical aspects of the AI ​​agent meet the standards. The certification unit can issue a certificate to the certified AI agent. The audit department audits the activities of AI agents certified by the certification department. For example, the audit department periodically audits whether the activities of AI agents meet the certification standards. The audit department can monitor the activity status of AI agents and provide guidance for improvement as needed. In this way, the AI ​​agent certification system according to the embodiment can efficiently operate the AI ​​agent certification system and improve the practicality and reliability of AI agents.

[0030] The Application Reception Department accepts applications. The Application Reception Department can accept applications in various formats, such as paper applications and online applications. Specifically, for paper applications, applicants mail or deliver the necessary documents in person, and a staff member at the Application Reception Department verifies the contents of the documents. For online applications, applicants submit their application through a dedicated web portal, and the system automatically accepts the application. The Application Reception Department verifies the application content and checks that all necessary documents are included. If there are any deficiencies, the department notifies the applicant and requests corrections. The Application Reception Department also verifies the applicant's identity and implements procedures to confirm that the applicant is legitimately qualified. This includes verifying the applicant's identification documents and past performance. Furthermore, the Application Reception Department stores the application content as digital data, making it accessible to subsequent examination and review departments. This allows the Application Reception Department to efficiently and accurately receive applications and smoothly transition them to the next process.

[0031] The Testing Department tests applications received by the Application Acceptance Department. For example, the Testing Department tests the skills and ethical aspects of AI agents. Specifically, it can conduct tests to evaluate the AI ​​agent's programming skills and problem-solving abilities. Programming skills tests involve the AI ​​agent writing code to perform a specific task and evaluating whether the code functions correctly. Problem-solving ability tests evaluate how the AI ​​agent approaches a given problem and finds a solution. The Testing Department can also conduct tests to evaluate the AI ​​agent's ethical judgment. Ethical judgment tests assess how the AI ​​agent makes decisions when faced with ethical dilemmas. For example, it verifies whether the AI ​​agent makes appropriate decisions from the perspective of privacy protection and fairness. The Testing Department meticulously records these test results and provides them to the Review Department. Furthermore, the Testing Department regularly reviews its test standards and methodologies to ensure they are up-to-date with the latest technologies and ethical standards. This allows the Testing Department to rigorously evaluate the skills and ethical aspects of AI agents and provide reliable test results.

[0032] The Review Department reviews the results of tests conducted by the Testing Department. For example, the Review Department evaluates the accuracy and reliability of the test results. Specifically, it examines the test results provided by the Testing Department in detail to confirm that the tests were conducted appropriately and that the results are accurate. The Review Department analyzes the test result data and uses statistical methods to evaluate the reliability of the results. Furthermore, the Review Department determines whether or not to certify the AI ​​agent based on the test results. The Review Department verifies whether the AI ​​agent meets the certification criteria and, if not, can request additional testing or improvements. In addition, the Review Department has established procedures to handle objections and requests for re-examination of test results, maintaining a fair and transparent review process. The Review Department assigns reviewers with specialized knowledge and experience to the review of test results to ensure high review quality. This allows the Review Department to guarantee the accuracy and reliability of test results and to conduct the AI ​​agent certification process fairly and transparently.

[0033] The Certification Department grants certification based on the results of the review conducted by the Examination Department. For example, the Certification Department verifies whether the AI ​​agent's skills and ethical aspects meet the standards and grants certification. Specifically, it makes a final confirmation that the AI ​​agent meets the certification standards based on the review results provided by the Examination Department. The Certification Department conducts a detailed evaluation of the AI ​​agent's skills and ethical aspects, and issues a certificate if the standards are met. The certificate includes information such as the AI ​​agent's certification number, certification date, and validity period of the certification. Along with issuing the certificate, the Certification Department can register and publish the information of the certified AI agent in its database. This allows for the widespread sharing of information on certified AI agents, contributing to improved reliability. Furthermore, the Certification Department regularly reviews its certification standards and procedures to keep them up-to-date with the latest technologies and ethical standards. This enables the Certification Department to rigorously evaluate the skills and ethical aspects of AI agents and provide highly reliable certification.

[0034] The Audit Department audits the activities of AI agents certified by the Certification Department. For example, the Audit Department periodically audits whether the activities of AI agents meet the certification standards. Specifically, it regularly checks the activity status of certified AI agents and verifies that they comply with the standards. The Audit Department analyzes the activity records and log data of AI agents and provides guidance for improvement if there are violations of the standards. The Audit Department may also conduct on-site inspections and interviews to confirm that the activities of AI agents are being carried out appropriately. In addition, the Audit Department prepares and periodically publishes reports on the activities of AI agents. This ensures that the activities of AI agents are monitored with transparency. Furthermore, based on the audit results, the Audit Department can revoke or suspend certification as necessary. This allows the Audit Department to monitor the activities of AI agents to always maintain high standards and ensure reliability and security.

[0035] The AI ​​agent certification system includes a robot command unit that integrates with a physical robot to verify the AI ​​agent's behavioral and conversational capabilities. The robot command unit allows the AI ​​agent to issue instructions to the physical robot, which then executes actions based on those instructions. For example, the robot command unit can allow the AI ​​agent to issue instructions to the robot, such as movement or object manipulation. The robot command unit evaluates the robot's actions to ensure the AI ​​agent's instructions are accurate and practical. For example, it evaluates the precision and practicality of the robot's actions. The robot command unit verifies that the robot operates accurately based on the AI ​​agent's instructions. This ensures that the AI ​​agent's instructions are accurate and practical. Some or all of the above-described processes in the robot command unit may be performed using or without a generative AI. For example, the robot command unit can input the AI ​​agent's instructions into a generative AI, which can then issue appropriate action instructions to the robot.

[0036] The AI ​​agent certification system includes a robot evaluation unit in which an autonomous examiner AI agent evaluates the robot's movements and verifies their accuracy and practicality. The robot evaluation unit evaluates whether the robot operates accurately based on the instructions of the AI ​​agent. For example, the robot evaluation unit evaluates the accuracy and practicality of the robot's movements. The robot evaluation unit verifies whether the robot operates accurately based on the instructions of the AI ​​agent. This ensures that the instructions of the AI ​​agent are accurate and practical. Some or all of the above-described processes in the robot evaluation unit may be performed using a generative AI, or they may be performed without a generative AI. For example, the robot evaluation unit can input the instructions of the AI ​​agent into a generative AI, and the generative AI can evaluate the robot's movements.

[0037] The AI ​​agent certification system includes a dialogue evaluation unit that uses generative AI to engage in dialogue and evaluates the dialogue capabilities of the AI ​​agent during that dialogue. The dialogue evaluation unit engages in dialogue with the AI ​​agent using generative AI and evaluates the dialogue capabilities of the AI ​​agent during that dialogue. For example, the dialogue evaluation unit evaluates the accuracy of the AI ​​agent's dialogue capabilities through the dialogue between the generative AI and the AI ​​agent. The dialogue evaluation unit analyzes the content of the dialogue between the generative AI and the AI ​​agent and evaluates the naturalness of the dialogue and the accuracy of the responses. This allows for confirmation of the accuracy of the AI ​​agent's dialogue capabilities. Some or all of the above-described processes in the dialogue evaluation unit may be performed using generative AI or without generative AI. For example, the dialogue evaluation unit can input the content of the dialogue with the AI ​​agent into the generative AI, and the generative AI can perform the dialogue evaluation.

[0038] The AI ​​agent certification system includes an instruction analysis unit in which a generating AI analyzes the instructions of the AI ​​agent and issues appropriate action instructions to a physical robot. The instruction analysis unit uses the generating AI to analyze the instructions of the AI ​​agent and issues appropriate action instructions to the physical robot. For example, the instruction analysis unit uses the generating AI to analyze the content of the instructions of the AI ​​agent and issues appropriate action instructions to the physical robot. The instruction analysis unit improves the accuracy of instructions by having the generating AI analyze the instructions of the AI ​​agent and issue accurate action instructions to the physical robot. This improves the accuracy of instructions to the physical robot. Some or all of the above processing in the instruction analysis unit may be performed using the generating AI or without using the generating AI. For example, the instruction analysis unit can input the content of the instructions of the AI ​​agent into the generating AI, and the generating AI can issue appropriate action instructions to the physical robot.

[0039] The application acceptance department can analyze past application data and select the optimal application acceptance method. For example, the application acceptance department can select the most efficient application acceptance method from past application data. Furthermore, the application acceptance department can analyze past application data and propose the optimal acceptance method based on applicant trends. In addition, the application acceptance department can select an acceptance method tailored to the applicant's needs based on past application data. Thus, by analyzing past application data, the optimal application acceptance method can be selected. Some or all of the above processes in the application acceptance department may be performed using AI, or not. For example, the application acceptance department can input past application data into AI, which can then select the optimal application acceptance method.

[0040] The application receiving unit can filter applications based on the applicant's current situation and areas of interest upon receipt. For example, the application receiving unit can prioritize accepting applications that are highly relevant based on the applicant's current situation. It can also filter appropriate applications based on the applicant's areas of interest. Furthermore, the application receiving unit can select the optimal application receiving method, taking into account the applicant's situation and areas of interest. This allows for the acceptance of appropriate applications by filtering based on the applicant's situation and areas of interest. Some or all of the above processing in the application receiving unit may be performed using AI or not. For example, the application receiving unit can input data on the applicant's situation and areas of interest into an AI, which can then filter appropriate applications.

[0041] The application receiving unit can prioritize the acceptance of highly relevant applications by considering the applicant's geographical location information when receiving applications. For example, the application receiving unit can prioritize the acceptance of highly relevant applications based on the applicant's geographical location information. The application receiving unit can also select the optimal application acceptance method by considering the applicant's geographical location information. Furthermore, the application receiving unit can filter appropriate applications based on the applicant's geographical location information. This allows for the priority acceptance of highly relevant applications by considering the applicant's geographical location information. Some or all of the above processing in the application receiving unit may be performed using AI or not. For example, the application receiving unit can input the applicant's geographical location information into AI, which can then prioritize the acceptance of highly relevant applications.

[0042] The application receiving department can analyze the applicant's social media activity upon receiving an application and accept relevant applications. For example, the application receiving department can analyze the applicant's social media activity and prioritize the acceptance of relevant applications. The application receiving department can also select the optimal application acceptance method based on the applicant's social media activity. Furthermore, the application receiving department can filter appropriate applications by considering the applicant's social media activity. This allows the application receiving department to accept relevant applications by analyzing the applicant's social media activity. Some or all of the above processing in the application receiving department may be performed using AI or not. For example, the application receiving department can input data on the applicant's social media activity into an AI, which can then accept relevant applications.

[0043] The testing unit can adjust the level of detail of the AI ​​agent during testing based on its importance. For example, the testing unit can conduct detailed testing for high-importance AI agents, and simplified testing for low-importance AI agents. Furthermore, the testing unit can adjust the level of detail of the test according to the importance of the AI ​​agent. This ensures that appropriate testing is provided by adjusting the level of detail according to the importance of the AI ​​agent. Some or all of the above processes in the testing unit may be performed using AI or not. For example, the testing unit can input AI agent importance data into an AI, which can then adjust the level of detail of the test.

[0044] The testing unit can apply different testing algorithms depending on the category of the AI ​​agent during testing. For example, the testing unit can select an appropriate testing algorithm depending on the category of the AI ​​agent. The testing unit can also customize the testing algorithm based on the category of the AI ​​agent. Furthermore, the testing unit can apply different testing algorithms depending on the category of the AI ​​agent. This improves the accuracy of the testing by applying the appropriate testing algorithm according to the category of the AI ​​agent. Some or all of the above processes in the testing unit may be performed using AI or not. For example, the testing unit can input AI agent category data into an AI, which can then select an appropriate testing algorithm.

[0045] The testing department can determine the priority of the AI ​​agent tests based on the submission timing during testing. For example, the testing department can prioritize testing AI agents that submit earlier. Conversely, the testing department can postpone testing AI agents that submit later. Furthermore, the testing department can determine the priority of the tests based on the submission timing of the AI ​​agents. This allows for more efficient testing by prioritizing tests based on the submission timing of the AI ​​agents. Some or all of the above processes in the testing department may be performed using AI or not. For example, the testing department can input AI agent submission timing data into an AI, which can then determine the priority of the tests.

[0046] The testing unit can adjust the order of testing based on the relevance of the AI ​​agents during testing. For example, the testing unit can prioritize testing highly relevant AI agents. It can also postpone testing less relevant AI agents. Furthermore, the testing unit can adjust the order of testing based on the relevance of the AI ​​agents. This allows for more efficient testing by adjusting the order of testing based on the relevance of the AI ​​agents. Some or all of the above processing in the testing unit may be performed using AI or not. For example, the testing unit can input AI agent relevance data into an AI, which can then adjust the order of testing.

[0047] The review department can improve the accuracy of its reviews by considering the interrelationships of AI agents during the review process. For example, the review department can analyze the interrelationships of AI agents to improve the accuracy of the review. The review department can also adjust the review criteria by considering the interrelationships of AI agents. Furthermore, the review department can improve the accuracy of the review based on the interrelationships of AI agents. In this way, the accuracy of the review is improved by considering the interrelationships of AI agents. Some or all of the above processes in the review department may be performed using AI or not. For example, the review department can input data on the interrelationships of AI agents into an AI, which can then improve the accuracy of the review.

[0048] The review department can conduct its review by considering the attribute information of the submitter provided by the AI ​​agent. For example, the review department can adjust the review criteria based on the submitter's attribute information. The review department can also improve the accuracy of the review by considering the submitter's attribute information. Furthermore, the review department can conduct the review based on the submitter's attribute information. This improves the accuracy of the review by considering the submitter's attribute information. Some or all of the above processes in the review department may be performed using AI or not. For example, the review department can input the submitter's attribute information data into the AI, which can then improve the accuracy of the review.

[0049] The review department can conduct reviews while considering the geographical distribution of AI agents. For example, the review department can adjust the review criteria based on the geographical distribution of AI agents. The review department can also improve the accuracy of the review by considering the geographical distribution of AI agents. Furthermore, the review department can conduct reviews based on the geographical distribution of AI agents. This improves the accuracy of the review by considering the geographical distribution of AI agents. Some or all of the above processes in the review department may be performed using AI or not. For example, the review department can input geographical distribution data of AI agents into an AI, which can then improve the accuracy of the review.

[0050] The review department can improve the accuracy of its review by referring to the AI ​​agent's relevant literature during the review process. For example, the review department can improve the accuracy of its review by referring to the AI ​​agent's relevant literature. The review department can also adjust its review criteria based on the AI ​​agent's relevant literature. Furthermore, the review department can improve the accuracy of its review by considering the AI ​​agent's relevant literature. As a result, the accuracy of the review is improved by referring to the AI ​​agent's relevant literature. Some or all of the above processes in the review department may be performed using AI or not. For example, the review department can input the AI ​​agent's relevant literature data into the AI, which can then improve the accuracy of its review.

[0051] The certification department can analyze the AI ​​agent's past activities during the certification process to select the optimal certification method. For example, the certification department can select the optimal certification method based on the AI ​​agent's past activity history. The certification department can also analyze the AI ​​agent's past activities and adjust the certification criteria. Furthermore, the certification department can customize the certification method considering the AI ​​agent's past activities. This allows the certification department to select the optimal certification method by analyzing the AI ​​agent's past activities. Some or all of the above processes in the certification department may be performed using AI or not. For example, the certification department can input the AI ​​agent's past activity data into an AI, which can then select the optimal certification method.

[0052] The certification unit can customize the certification process based on the current status of the AI ​​agent during the certification process. For example, the certification unit can customize the certification process based on the current status of the AI ​​agent. The certification unit can also adjust the certification criteria considering the current status of the AI ​​agent. Furthermore, the certification unit can select a certification method based on the current status of the AI ​​agent. This allows for more appropriate certification by customizing the certification process based on the current status of the AI ​​agent. Some or all of the above processes in the certification unit may be performed using AI or not. For example, the certification unit can input the current status data of the AI ​​agent into the AI, and the AI ​​can customize the certification process.

[0053] The certification unit can select the optimal certification method by considering the geographical location information of the AI ​​agent during the certification process. For example, the certification unit can select the optimal certification method based on the geographical location information of the AI ​​agent. The certification unit can also adjust the certification criteria by considering the geographical location information of the AI ​​agent. Furthermore, the certification unit can customize the certification means based on the geographical location information of the AI ​​agent. This allows for the selection of the optimal certification method by considering the geographical location information of the AI ​​agent. Some or all of the above-described processes in the certification unit may be performed using AI or not. For example, the certification unit can input the geographical location data of the AI ​​agent into the AI, and the AI ​​can select the optimal certification method.

[0054] The certification department can analyze the social media activity of an AI agent during the certification process and propose certification methods. For example, the certification department can analyze the social media activity of an AI agent and propose the optimal certification method. The certification department can also adjust the certification criteria based on the social media activity of the AI ​​agent. Furthermore, the certification department can customize the certification methods considering the social media activity of the AI ​​agent. This allows the certification department to propose the optimal certification method by analyzing the social media activity of the AI ​​agent. Some or all of the above processes in the certification department may be performed using AI or not. For example, the certification department can input the social media activity data of the AI ​​agent into an AI, which can then propose the optimal certification method.

[0055] The audit department can analyze the AI ​​agent's past activities during an audit to select the optimal audit method. For example, the audit department can select the optimal audit method based on the AI ​​agent's past activity history. The audit department can also analyze the AI ​​agent's past activities and adjust the audit standards. Furthermore, the audit department can customize the audit method considering the AI ​​agent's past activities. This allows the audit department to select the optimal audit method by analyzing the AI ​​agent's past activities. Some or all of the above processes in the audit department may be performed using AI or not. For example, the audit department can input the AI ​​agent's past activity data into the AI, which can then select the optimal audit method.

[0056] The audit department can customize audit methods based on the current status of the AI ​​agent during an audit. For example, the audit department can customize audit methods based on the current status of the AI ​​agent. The audit department can also adjust audit criteria considering the current status of the AI ​​agent. Furthermore, the audit department can select audit methods based on the current status of the AI ​​agent. This allows for more appropriate audits by customizing audit methods based on the current status of the AI ​​agent. Some or all of the above processes in the audit department may be performed using AI or not. For example, the audit department can input current status data of the AI ​​agent into the AI, and the AI ​​can customize the audit methods.

[0057] The audit department can select the optimal audit method during an audit by considering the geographical location information of the AI ​​agent. For example, the audit department can select the optimal audit method based on the geographical location information of the AI ​​agent. The audit department can also adjust the audit criteria by considering the geographical location information of the AI ​​agent. Furthermore, the audit department can customize the audit methods based on the geographical location information of the AI ​​agent. This allows for the selection of the optimal audit method by considering the geographical location information of the AI ​​agent. Some or all of the above processes performed by the audit department may be carried out using AI or not. For example, the audit department can input the geographical location data of the AI ​​agent into the AI, and the AI ​​can select the optimal audit method.

[0058] The audit department can analyze the social media activity of AI agents during an audit and propose audit methods. For example, the audit department can analyze the social media activity of AI agents and propose the most suitable audit methods. The audit department can also adjust audit standards based on the social media activity of AI agents. Furthermore, the audit department can customize audit methods considering the social media activity of AI agents. This allows the audit department to propose the most suitable audit methods by analyzing the social media activity of AI agents. Some or all of the above processes performed by the audit department may be carried out using AI or not. For example, the audit department can input the social media activity data of AI agents into an AI, which can then propose the most suitable audit methods.

[0059] The robot instruction unit can analyze the AI ​​agent's past instruction history to select the optimal instruction method when giving instructions to the robot. For example, the robot instruction unit can select the optimal instruction method based on the AI ​​agent's past instruction history. The robot instruction unit can also analyze the AI ​​agent's past instruction history and adjust the instruction criteria. Furthermore, the robot instruction unit can customize the instruction method considering the AI ​​agent's past instruction history. This allows the optimal instruction method to be selected by analyzing the AI ​​agent's past instruction history. Some or all of the above processes in the robot instruction unit may be performed using AI or not. For example, the robot instruction unit can input the AI ​​agent's past instruction history data into the AI, and the AI ​​can select the optimal instruction method.

[0060] The robot instruction unit can customize the instruction methods based on the current status of the AI ​​agent when giving instructions to the robot. For example, the robot instruction unit customizes the instruction methods based on the current status of the AI ​​agent. The robot instruction unit can also adjust the instruction criteria considering the current status of the AI ​​agent. Furthermore, the robot instruction unit can select an instruction method based on the current status of the AI ​​agent. This allows for more appropriate instructions by customizing the instruction methods based on the current status of the AI ​​agent. Some or all of the above processes in the robot instruction unit may be performed using AI or not. For example, the robot instruction unit can input the current status data of the AI ​​agent into the AI, and the AI ​​can customize the instruction methods.

[0061] The robot instruction unit can select the optimal instruction method when giving instructions to the robot, taking into account the geographical location information of the AI ​​agent. For example, the robot instruction unit can select the optimal instruction method based on the geographical location information of the AI ​​agent. The robot instruction unit can also adjust the instruction criteria, taking into account the geographical location information of the AI ​​agent. Furthermore, the robot instruction unit can customize the instruction means based on the geographical location information of the AI ​​agent. This allows for the selection of the optimal instruction method by considering the geographical location information of the AI ​​agent. Some or all of the above processing in the robot instruction unit may be performed using AI, or not using AI. For example, the robot instruction unit can input the geographical location data of the AI ​​agent into the AI, and the AI ​​can select the optimal instruction method.

[0062] The robot instruction unit can analyze the AI ​​agent's social media activity and propose instruction methods when issuing robot instructions. For example, the robot instruction unit can analyze the AI ​​agent's social media activity and propose the optimal instruction method. The robot instruction unit can also adjust instruction criteria based on the AI ​​agent's social media activity. Furthermore, the robot instruction unit can customize instruction methods considering the AI ​​agent's social media activity. This allows the optimal instruction method to be proposed by analyzing the AI ​​agent's social media activity. Some or all of the above processing in the robot instruction unit may be performed using AI or not. For example, the robot instruction unit can input the AI ​​agent's social media activity data into the AI, which can then propose the optimal instruction method.

[0063] The robot evaluation unit can select the optimal evaluation method by analyzing the past evaluation history of the AI ​​agent during robot evaluation. For example, the robot evaluation unit selects the optimal evaluation method based on the past evaluation history of the AI ​​agent. The robot evaluation unit can also analyze the past evaluation history of the AI ​​agent and adjust the evaluation criteria. Furthermore, the robot evaluation unit can customize the evaluation method by considering the past evaluation history of the AI ​​agent. This allows the robot evaluation unit to select the optimal evaluation method by analyzing the past evaluation history of the AI ​​agent. Some or all of the above processes in the robot evaluation unit may be performed using AI or not. For example, the robot evaluation unit can input the past evaluation history data of the AI ​​agent into the AI, and the AI ​​can select the optimal evaluation method.

[0064] The robot evaluation unit can customize the evaluation methods based on the current status of the AI ​​agent during robot evaluation. For example, the robot evaluation unit customizes the evaluation methods based on the current status of the AI ​​agent. The robot evaluation unit can also adjust the evaluation criteria considering the current status of the AI ​​agent. Furthermore, the robot evaluation unit can select an evaluation method based on the current status of the AI ​​agent. This allows for a more appropriate evaluation by customizing the evaluation methods based on the current status of the AI ​​agent. Some or all of the above processes in the robot evaluation unit may be performed using AI or not. For example, the robot evaluation unit can input the current status data of the AI ​​agent into the AI, and the AI ​​can customize the evaluation methods.

[0065] The robot evaluation unit can select the optimal evaluation method when evaluating a robot, taking into account the geographical location information of the AI ​​agent. For example, the robot evaluation unit can select the optimal evaluation method based on the geographical location information of the AI ​​agent. The robot evaluation unit can also adjust the evaluation criteria, taking into account the geographical location information of the AI ​​agent. Furthermore, the robot evaluation unit can customize the evaluation means based on the geographical location information of the AI ​​agent. This allows for the selection of the optimal evaluation method by considering the geographical location information of the AI ​​agent. Some or all of the above-described processes in the robot evaluation unit may be performed using AI, or they may not be performed using AI. For example, the robot evaluation unit can input the geographical location data of the AI ​​agent into the AI, and the AI ​​can select the optimal evaluation method.

[0066] The robot evaluation unit can analyze the social media activity of an AI agent during robot evaluation and propose evaluation methods. For example, the robot evaluation unit can analyze the social media activity of an AI agent and propose the optimal evaluation method. The robot evaluation unit can also adjust the evaluation criteria based on the social media activity of the AI ​​agent. Furthermore, the robot evaluation unit can customize the evaluation method by taking into account the social media activity of the AI ​​agent. This allows the robot evaluation unit to propose the optimal evaluation method by analyzing the social media activity of the AI ​​agent. Some or all of the above processing in the robot evaluation unit may be performed using AI or not. For example, the robot evaluation unit can input the social media activity data of the AI ​​agent into the AI, and the AI ​​can propose the optimal evaluation method.

[0067] The dialogue evaluation unit can analyze the AI ​​agent's past dialogue history during dialogue evaluation to select the optimal evaluation method. For example, the dialogue evaluation unit selects the optimal evaluation method based on the AI ​​agent's past dialogue history. The dialogue evaluation unit can also analyze the AI ​​agent's past dialogue history and adjust the evaluation criteria. Furthermore, the dialogue evaluation unit can customize the evaluation method by considering the AI ​​agent's past dialogue history. This allows the optimal evaluation method to be selected by analyzing the AI ​​agent's past dialogue history. Some or all of the above processing in the dialogue evaluation unit may be performed using a generative AI, or it may be performed without a generative AI. For example, the dialogue evaluation unit can input the AI ​​agent's past dialogue history data into a generative AI, which can then select the optimal evaluation method.

[0068] The dialogue evaluation unit can customize the evaluation methods based on the current state of the AI ​​agent during dialogue evaluation. For example, the dialogue evaluation unit customizes the evaluation methods based on the current state of the AI ​​agent. The dialogue evaluation unit can also adjust the evaluation criteria considering the current state of the AI ​​agent. Furthermore, the dialogue evaluation unit can select an evaluation method based on the current state of the AI ​​agent. This allows for a more appropriate evaluation by customizing the evaluation methods based on the current state of the AI ​​agent. Some or all of the above processing in the dialogue evaluation unit may be performed using a generating AI, or it may be performed without a generating AI. For example, the dialogue evaluation unit can input the current state data of the AI ​​agent into a generating AI, and the generating AI can customize the evaluation methods.

[0069] The dialogue evaluation unit can select the optimal evaluation method during dialogue evaluation by considering the geographical location information of the AI ​​agent. For example, the dialogue evaluation unit selects the optimal evaluation method based on the geographical location information of the AI ​​agent. The dialogue evaluation unit can also adjust the evaluation criteria by considering the geographical location information of the AI ​​agent. Furthermore, the dialogue evaluation unit can customize the evaluation means based on the geographical location information of the AI ​​agent. This allows for the selection of the optimal evaluation method by considering the geographical location information of the AI ​​agent. Some or all of the above processing in the dialogue evaluation unit may be performed using a generating AI, or it may be performed without using a generating AI. For example, the dialogue evaluation unit can input the geographical location information data of the AI ​​agent into a generating AI, and the generating AI can select the optimal evaluation method.

[0070] The dialogue evaluation unit can analyze the AI ​​agent's social media activity during dialogue evaluation and propose evaluation methods. For example, the dialogue evaluation unit can analyze the AI ​​agent's social media activity and propose the optimal evaluation method. The dialogue evaluation unit can also adjust the evaluation criteria based on the AI ​​agent's social media activity. Furthermore, the dialogue evaluation unit can customize the evaluation method by taking the AI ​​agent's social media activity into consideration. This allows the optimal evaluation method to be proposed by analyzing the AI ​​agent's social media activity. Some or all of the above processing in the dialogue evaluation unit may be performed using a generative AI, or it may be performed without a generative AI. For example, the dialogue evaluation unit can input the AI ​​agent's social media activity data into a generative AI, and the generative AI can propose the optimal evaluation method.

[0071] The instruction analysis unit can analyze the AI ​​agent's past instruction history and select the optimal analysis method during instruction analysis. For example, the instruction analysis unit selects the optimal analysis method based on the AI ​​agent's past instruction history. The instruction analysis unit can also analyze the AI ​​agent's past instruction history and adjust the analysis criteria. Furthermore, the instruction analysis unit can customize the analysis method considering the AI ​​agent's past instruction history. This allows the optimal analysis method to be selected by analyzing the AI ​​agent's past instruction history. Some or all of the above processing in the instruction analysis unit may be performed using a generating AI, or it may be performed without using a generating AI. For example, the instruction analysis unit can input the AI ​​agent's past instruction history data into a generating AI, which can then select the optimal analysis method.

[0072] The instruction analysis unit can customize the analysis methods based on the current status of the AI ​​agent during instruction analysis. For example, the instruction analysis unit customizes the analysis methods based on the current status of the AI ​​agent. The instruction analysis unit can also adjust the analysis criteria considering the current status of the AI ​​agent. Furthermore, the instruction analysis unit can select an analysis method based on the current status of the AI ​​agent. This allows for more appropriate analysis by customizing the analysis methods based on the current status of the AI ​​agent. Some or all of the above processing in the instruction analysis unit may be performed using a generating AI, or it may be performed without using a generating AI. For example, the instruction analysis unit can input the current status data of the AI ​​agent into a generating AI, and the generating AI can customize the analysis methods.

[0073] The instruction analysis unit can select the optimal analysis method by considering the geographical location information of the AI ​​agent during instruction analysis. For example, the instruction analysis unit selects the optimal analysis method based on the geographical location information of the AI ​​agent. The instruction analysis unit can also adjust the analysis criteria by considering the geographical location information of the AI ​​agent. Furthermore, the instruction analysis unit can customize the analysis means based on the geographical location information of the AI ​​agent. This allows for the selection of the optimal analysis method by considering the geographical location information of the AI ​​agent. Some or all of the above-described processes in the instruction analysis unit may be performed using a generating AI, or they may be performed without using a generating AI. For example, the instruction analysis unit can input the geographical location information data of the AI ​​agent into a generating AI, and the generating AI can select the optimal analysis method.

[0074] The instruction analysis unit can analyze the AI ​​agent's social media activities during instruction analysis and propose analysis methods. For example, the instruction analysis unit can analyze the AI ​​agent's social media activities and propose the optimal analysis method. The instruction analysis unit can also adjust the analysis criteria based on the AI ​​agent's social media activities. Furthermore, the instruction analysis unit can customize the analysis method considering the AI ​​agent's social media activities. This allows the optimal analysis method to be proposed by analyzing the AI ​​agent's social media activities. Some or all of the above processing in the instruction analysis unit may be performed using a generating AI, or it may be performed without using a generating AI. For example, the instruction analysis unit can input the AI ​​agent's social media activity data into a generating AI, and the generating AI can propose the optimal analysis method.

[0075] The system according to the embodiment is not limited to the example described above, and various modifications are possible, for example, as follows.

[0076] The AI ​​agent certification system allows the application processing department to analyze an applicant's past application history and propose the optimal application processing method. For example, applicants who have previously requested expedited processing will have their applications prioritized. Furthermore, applicants who have submitted complex applications in the past can be provided with detailed guidelines. The system can also analyze applicant trends based on their past application history and customize the optimal application processing method. This allows for more appropriate application processing by considering the applicant's past application history.

[0077] The AI ​​agent certification system allows the examination department to analyze the learning history of AI agents and provide optimal exam content. For example, an AI agent that has achieved high scores in a particular field in the past will be given an advanced exam related to that field. Similarly, an AI agent that has struggled in a particular area in the past will be given an exam focused on that area. Furthermore, the exam content can be customized based on the learning history to promote the AI ​​agent's growth. This allows for the provision of more appropriate exams by considering the AI ​​agent's learning history.

[0078] The AI ​​agent certification system allows the review department to implement peer evaluation of AI agents, thereby improving the accuracy of the review process. For example, it can collect feedback from other AI agents and use it as a reference for the review. Furthermore, based on peer evaluation, it can analyze the strengths and weaknesses of AI agents and adjust the review criteria. It can also utilize peer evaluation to assess the collaborative relationships between AI agents. In this way, implementing peer evaluation of AI agents improves the accuracy of the review process.

[0079] The AI ​​agent certification system allows the certification department to publish the performance of AI agents, thereby increasing transparency. For example, the performance of certified AI agents can be made public online, allowing other applicants and stakeholders to view it. Furthermore, the certification criteria can be reviewed based on the performance, and stricter standards can be set. In addition, the reliability of AI agents can be improved through the publication of performance. In this way, by disclosing the performance of AI agents, the transparency of the certification process is increased and the reliability is improved.

[0080] The AI ​​agent certification system allows the audit department to monitor AI agent activities in real time, enabling rapid response. For example, it can monitor AI agent activity logs in real time and respond immediately if an anomaly occurs. Furthermore, based on real-time monitoring, the AI ​​agent's performance can be evaluated, and improvement guidance can be provided as needed. In addition, real-time monitoring can increase the transparency of AI agent activities. This results in faster response times and improved transparency through real-time monitoring of AI agent activities.

[0081] The following briefly describes the processing flow for example form 1.

[0082] Step 1: The application receiving department accepts the application. The application receiving department can accept applications in various formats, such as paper applications and online applications. Step 2: The Testing Department tests the applications received by the Application Acceptance Department. The Testing Department may, for example, test the skills and ethical aspects of the AI ​​agent. The Testing Department may conduct tests to evaluate the AI ​​agent's programming skills and problem-solving abilities. The Testing Department may also conduct tests to evaluate the AI ​​agent's ethical judgment abilities. Step 3: The Review Department reviews the results of the tests conducted by the Test Department. The Review Department evaluates, for example, the accuracy and reliability of the test results. Based on the test results, the Review Department decides whether or not to certify the AI ​​agent. Step 4: The Certification Department certifies the AI ​​agents based on the results of the review conducted by the Examination Department. The Certification Department certifies, for example, whether the AI ​​agents' skills and ethical aspects meet the standards. The Certification Department can issue a certificate to the certified AI agents. Step 5: The Audit Department audits the activities of AI agents certified by the Certification Department. For example, the Audit Department periodically audits whether the AI ​​agents' activities meet the certification standards. The Audit Department can monitor the AI ​​agents' activities and provide guidance for improvement as needed.

[0083] (Example of form 2) An AI agent certification system according to an embodiment of the present invention organizes a group of AI agents for operating an AI agent certification system and provides an agency that leads the agents responsible for each task. In the AI ​​agent certification system, the application acceptance department accepts certification applications for AI agents, the examination department tests the skills and ethical aspects of the applied AI agents, the review department reviews the test results, the certification department grants certification, and the audit department audits the activities of the certified AI agents. Furthermore, to verify the behavioral and conversational capabilities of the AI ​​agents, collaboration with a physical robot is introduced. For example, the AI ​​agent gives instructions to the physical robot, and the robot performs actions based on those instructions. The autonomous examiner AI agent evaluates the robot's actions and verifies their accuracy and practicality. This ensures that the AI ​​agent's instructions are accurate and practical. Furthermore, the autonomous examiner AI agent autonomously performs evaluation tasks and provides feedback. For example, it uses generative AI to engage in dialogue and evaluates the AI ​​agent's conversational ability during that dialogue. In addition, the generative AI analyzes the AI ​​agent's instructions and gives appropriate action instructions to the physical robot, thereby evaluating the accuracy of instruction understanding and execution. This mechanism allows for the efficient operation of the AI ​​agent certification system and improves the practicality and reliability of AI agents. For example, by using physical robots to evaluate the accuracy and practicality of AI agent execution and providing accurate and efficient feedback, the practicality and reliability of AI agents are enhanced. As a result, the AI ​​agent certification system can efficiently operate the AI ​​agent certification system and improve the practicality and reliability of AI agents.

[0084] The AI ​​agent certification system according to this embodiment comprises an application acceptance unit, an examination unit, an evaluation unit, a certification unit, and an audit unit. The application acceptance unit accepts applications. The application acceptance unit can accept applications in various forms, such as paper applications and online applications. The examination unit tests the applications accepted by the application acceptance unit. The examination unit tests, for example, the skills and ethical aspects of the AI ​​agent. The examination unit can conduct tests to evaluate the programming skills and problem-solving abilities of the AI ​​agent. The examination unit can also conduct tests to evaluate the ethical judgment abilities of the AI ​​agent. The evaluation unit evaluates the results of the tests conducted by the examination unit. The evaluation unit evaluates, for example, the accuracy and reliability of the test results. The evaluation unit determines whether or not to certify the AI ​​agent based on the test results. The certification unit certifies the AI ​​agent based on the results of the evaluation conducted by the evaluation unit. The certification unit certifies the AI ​​agent, for example, by confirming whether the skills and ethical aspects of the AI ​​agent meet the standards. The certification unit can issue a certificate to the certified AI agent. The audit department audits the activities of AI agents certified by the certification department. For example, the audit department periodically audits whether the activities of AI agents meet the certification standards. The audit department can monitor the activity status of AI agents and provide guidance for improvement as needed. In this way, the AI ​​agent certification system according to the embodiment can efficiently operate the AI ​​agent certification system and improve the practicality and reliability of AI agents.

[0085] The Application Reception Department accepts applications. The Application Reception Department can accept applications in various formats, such as paper applications and online applications. Specifically, for paper applications, applicants mail or deliver the necessary documents in person, and a staff member at the Application Reception Department verifies the contents of the documents. For online applications, applicants submit their application through a dedicated web portal, and the system automatically accepts the application. The Application Reception Department verifies the application content and checks that all necessary documents are included. If there are any deficiencies, the department notifies the applicant and requests corrections. The Application Reception Department also verifies the applicant's identity and implements procedures to confirm that the applicant is legitimately qualified. This includes verifying the applicant's identification documents and past performance. Furthermore, the Application Reception Department stores the application content as digital data, making it accessible to subsequent examination and review departments. This allows the Application Reception Department to efficiently and accurately receive applications and smoothly transition them to the next process.

[0086] The Testing Department tests applications received by the Application Acceptance Department. For example, the Testing Department tests the skills and ethical aspects of AI agents. Specifically, it can conduct tests to evaluate the AI ​​agent's programming skills and problem-solving abilities. Programming skills tests involve the AI ​​agent writing code to perform a specific task and evaluating whether the code functions correctly. Problem-solving ability tests evaluate how the AI ​​agent approaches a given problem and finds a solution. The Testing Department can also conduct tests to evaluate the AI ​​agent's ethical judgment. Ethical judgment tests assess how the AI ​​agent makes decisions when faced with ethical dilemmas. For example, it verifies whether the AI ​​agent makes appropriate decisions from the perspective of privacy protection and fairness. The Testing Department meticulously records these test results and provides them to the Review Department. Furthermore, the Testing Department regularly reviews its test standards and methodologies to ensure they are up-to-date with the latest technologies and ethical standards. This allows the Testing Department to rigorously evaluate the skills and ethical aspects of AI agents and provide reliable test results.

[0087] The Review Department reviews the results of tests conducted by the Testing Department. For example, the Review Department evaluates the accuracy and reliability of the test results. Specifically, it examines the test results provided by the Testing Department in detail to confirm that the tests were conducted appropriately and that the results are accurate. The Review Department analyzes the test result data and uses statistical methods to evaluate the reliability of the results. Furthermore, the Review Department determines whether or not to certify the AI ​​agent based on the test results. The Review Department verifies whether the AI ​​agent meets the certification criteria and, if not, can request additional testing or improvements. In addition, the Review Department has established procedures to handle objections and requests for re-examination of test results, maintaining a fair and transparent review process. The Review Department assigns reviewers with specialized knowledge and experience to the review of test results to ensure high review quality. This allows the Review Department to guarantee the accuracy and reliability of test results and to conduct the AI ​​agent certification process fairly and transparently.

[0088] The Certification Department grants certification based on the results of the review conducted by the Examination Department. For example, the Certification Department verifies whether the AI ​​agent's skills and ethical aspects meet the standards and grants certification. Specifically, it makes a final confirmation that the AI ​​agent meets the certification standards based on the review results provided by the Examination Department. The Certification Department conducts a detailed evaluation of the AI ​​agent's skills and ethical aspects, and issues a certificate if the standards are met. The certificate includes information such as the AI ​​agent's certification number, certification date, and validity period of the certification. Along with issuing the certificate, the Certification Department can register and publish the information of the certified AI agent in its database. This allows for the widespread sharing of information on certified AI agents, contributing to improved reliability. Furthermore, the Certification Department regularly reviews its certification standards and procedures to keep them up-to-date with the latest technologies and ethical standards. This enables the Certification Department to rigorously evaluate the skills and ethical aspects of AI agents and provide highly reliable certification.

[0089] The Audit Department audits the activities of AI agents certified by the Certification Department. For example, the Audit Department periodically audits whether the activities of AI agents meet the certification standards. Specifically, it regularly checks the activity status of certified AI agents and verifies that they comply with the standards. The Audit Department analyzes the activity records and log data of AI agents and provides guidance for improvement if there are violations of the standards. The Audit Department may also conduct on-site inspections and interviews to confirm that the activities of AI agents are being carried out appropriately. In addition, the Audit Department prepares and periodically publishes reports on the activities of AI agents. This ensures that the activities of AI agents are monitored with transparency. Furthermore, based on the audit results, the Audit Department can revoke or suspend certification as necessary. This allows the Audit Department to monitor the activities of AI agents to always maintain high standards and ensure reliability and security.

[0090] The AI ​​agent certification system includes a robot command unit that integrates with a physical robot to verify the AI ​​agent's behavioral and conversational capabilities. The robot command unit allows the AI ​​agent to issue instructions to the physical robot, which then executes actions based on those instructions. For example, the robot command unit can allow the AI ​​agent to issue instructions to the robot, such as movement or object manipulation. The robot command unit evaluates the robot's actions to ensure the AI ​​agent's instructions are accurate and practical. For example, it evaluates the precision and practicality of the robot's actions. The robot command unit verifies that the robot operates accurately based on the AI ​​agent's instructions. This ensures that the AI ​​agent's instructions are accurate and practical. Some or all of the above-described processes in the robot command unit may be performed using or without a generative AI. For example, the robot command unit can input the AI ​​agent's instructions into a generative AI, which can then issue appropriate action instructions to the robot.

[0091] The AI ​​agent certification system includes a robot evaluation unit in which an autonomous examiner AI agent evaluates the robot's movements and verifies their accuracy and practicality. The robot evaluation unit evaluates whether the robot operates accurately based on the instructions of the AI ​​agent. For example, the robot evaluation unit evaluates the accuracy and practicality of the robot's movements. The robot evaluation unit verifies whether the robot operates accurately based on the instructions of the AI ​​agent. This ensures that the instructions of the AI ​​agent are accurate and practical. Some or all of the above-described processes in the robot evaluation unit may be performed using a generative AI, or they may be performed without a generative AI. For example, the robot evaluation unit can input the instructions of the AI ​​agent into a generative AI, and the generative AI can evaluate the robot's movements.

[0092] The AI ​​agent certification system includes a dialogue evaluation unit that uses generative AI to engage in dialogue and evaluates the dialogue capabilities of the AI ​​agent during that dialogue. The dialogue evaluation unit engages in dialogue with the AI ​​agent using generative AI and evaluates the dialogue capabilities of the AI ​​agent during that dialogue. For example, the dialogue evaluation unit evaluates the accuracy of the AI ​​agent's dialogue capabilities through the dialogue between the generative AI and the AI ​​agent. The dialogue evaluation unit analyzes the content of the dialogue between the generative AI and the AI ​​agent and evaluates the naturalness of the dialogue and the accuracy of the responses. This allows for confirmation of the accuracy of the AI ​​agent's dialogue capabilities. Some or all of the above-described processes in the dialogue evaluation unit may be performed using generative AI or without generative AI. For example, the dialogue evaluation unit can input the content of the dialogue with the AI ​​agent into the generative AI, and the generative AI can perform the dialogue evaluation.

[0093] The AI ​​agent certification system includes an instruction analysis unit in which a generating AI analyzes the instructions of the AI ​​agent and issues appropriate action instructions to a physical robot. The instruction analysis unit uses the generating AI to analyze the instructions of the AI ​​agent and issues appropriate action instructions to the physical robot. For example, the instruction analysis unit uses the generating AI to analyze the content of the instructions of the AI ​​agent and issues appropriate action instructions to the physical robot. The instruction analysis unit improves the accuracy of instructions by having the generating AI analyze the instructions of the AI ​​agent and issue accurate action instructions to the physical robot. This improves the accuracy of instructions to the physical robot. Some or all of the above processing in the instruction analysis unit may be performed using the generating AI or without using the generating AI. For example, the instruction analysis unit can input the content of the instructions of the AI ​​agent into the generating AI, and the generating AI can issue appropriate action instructions to the physical robot.

[0094] The application acceptance unit can estimate the user's emotions and adjust the timing of application acceptance based on the estimated emotions. For example, if the user is stressed, the application acceptance unit can delay the application acceptance and wait until the user is relaxed. Conversely, if the user is in a hurry, the application acceptance unit can expedite the application acceptance to provide a quick response. Furthermore, if the user is anxious, the application acceptance unit can adjust the timing of application acceptance to provide reassurance. This allows for more appropriate application acceptance by adjusting the timing of application acceptance according to the user's emotions. Emotion estimation is achieved using an emotion estimation function, such as an emotion engine or generative AI. Generative AI may be, but is not limited to, text generation AI (e.g., LLM) or multimodal generation AI. Some or all of the above-described processes in the application acceptance unit may be performed using AI or not. For example, the application acceptance unit can input user emotion data into a generative AI, which can then perform emotion estimation.

[0095] The application acceptance department can analyze past application data and select the optimal application acceptance method. For example, the application acceptance department can select the most efficient application acceptance method from past application data. Furthermore, the application acceptance department can analyze past application data and propose the optimal acceptance method based on applicant trends. In addition, the application acceptance department can select an acceptance method tailored to the applicant's needs based on past application data. Thus, by analyzing past application data, the optimal application acceptance method can be selected. Some or all of the above processes in the application acceptance department may be performed using AI, or not. For example, the application acceptance department can input past application data into AI, which can then select the optimal application acceptance method.

[0096] The application receiving unit can filter applications based on the applicant's current situation and areas of interest upon receipt. For example, the application receiving unit can prioritize accepting applications that are highly relevant based on the applicant's current situation. It can also filter appropriate applications based on the applicant's areas of interest. Furthermore, the application receiving unit can select the optimal application receiving method, taking into account the applicant's situation and areas of interest. This allows for the acceptance of appropriate applications by filtering based on the applicant's situation and areas of interest. Some or all of the above processing in the application receiving unit may be performed using AI or not. For example, the application receiving unit can input data on the applicant's situation and areas of interest into an AI, which can then filter appropriate applications.

[0097] The application receiving unit can estimate the user's emotions and determine the priority of applications to be accepted based on the estimated emotions. For example, if the user is stressed, the application receiving unit can lower the priority of the application. Conversely, if the user is relaxed, the application receiving unit can raise the priority and respond quickly. Furthermore, if the user is in a hurry, the application receiving unit can raise the priority and respond quickly. This allows for more appropriate application acceptance by determining the priority of applications according to the user's emotions. Emotion estimation is achieved using an emotion estimation function, such as an emotion engine or generative AI. Generative AI may be, but is not limited to, text generation AI (e.g., LLM) or multimodal generation AI. Some or all of the above processing in the application receiving unit may be performed using AI or not. For example, the application receiving unit can input user emotion data into a generative AI, which can then estimate the emotions.

[0098] The application receiving unit can prioritize the acceptance of highly relevant applications by considering the applicant's geographical location information when receiving applications. For example, the application receiving unit can prioritize the acceptance of highly relevant applications based on the applicant's geographical location information. The application receiving unit can also select the optimal application acceptance method by considering the applicant's geographical location information. Furthermore, the application receiving unit can filter appropriate applications based on the applicant's geographical location information. This allows for the priority acceptance of highly relevant applications by considering the applicant's geographical location information. Some or all of the above processing in the application receiving unit may be performed using AI or not. For example, the application receiving unit can input the applicant's geographical location information into AI, which can then prioritize the acceptance of highly relevant applications.

[0099] The application receiving department can analyze the applicant's social media activity upon receiving an application and accept relevant applications. For example, the application receiving department can analyze the applicant's social media activity and prioritize the acceptance of relevant applications. The application receiving department can also select the optimal application acceptance method based on the applicant's social media activity. Furthermore, the application receiving department can filter appropriate applications by considering the applicant's social media activity. This allows the application receiving department to accept relevant applications by analyzing the applicant's social media activity. Some or all of the above processing in the application receiving department may be performed using AI or not. For example, the application receiving department can input data on the applicant's social media activity into an AI, which can then accept relevant applications.

[0100] The testing unit can estimate the user's emotions and adjust the presentation of the test based on the estimated emotions. For example, if the user is nervous, the testing unit can provide a simple and highly visual test. If the user is relaxed, the testing unit can also provide a test with more detailed information. Furthermore, if the user is in a hurry, the testing unit can provide a concise test. By adjusting the presentation of the test according to the user's emotions, a more appropriate test can be provided. Emotion estimation is achieved using an emotion estimation function, such as an emotion engine or generative AI. Generative AI is, but is not limited to, text generation AI (e.g., LLM) or multimodal generation AI. Some or all of the above processing in the testing unit may be performed using AI or not. For example, the testing unit can input user emotion data into a generative AI, which can then perform emotion estimation.

[0101] The testing unit can adjust the level of detail of the AI ​​agent during testing based on its importance. For example, the testing unit can conduct detailed testing for high-importance AI agents, and simplified testing for low-importance AI agents. Furthermore, the testing unit can adjust the level of detail of the test according to the importance of the AI ​​agent. This ensures that appropriate testing is provided by adjusting the level of detail according to the importance of the AI ​​agent. Some or all of the above processes in the testing unit may be performed using AI or not. For example, the testing unit can input AI agent importance data into an AI, which can then adjust the level of detail of the test.

[0102] The testing unit can apply different testing algorithms depending on the category of the AI ​​agent during testing. For example, the testing unit can select an appropriate testing algorithm depending on the category of the AI ​​agent. The testing unit can also customize the testing algorithm based on the category of the AI ​​agent. Furthermore, the testing unit can apply different testing algorithms depending on the category of the AI ​​agent. This improves the accuracy of the testing by applying the appropriate testing algorithm according to the category of the AI ​​agent. Some or all of the above processes in the testing unit may be performed using AI or not. For example, the testing unit can input AI agent category data into an AI, which can then select an appropriate testing algorithm.

[0103] The testing unit can estimate the user's emotions and adjust the length of the test based on the estimated emotions. For example, if the user is nervous, the testing unit can provide a shorter test. If the user is relaxed, the testing unit can also provide a longer test. Furthermore, if the user is in a hurry, the testing unit can provide a short, concise test. By adjusting the length of the test according to the user's emotions, a more appropriate test can be provided. Emotion estimation is achieved using an emotion estimation function, such as an emotion engine or generative AI. Generative AI is, but is not limited to, text generation AI (e.g., LLM) or multimodal generation AI. Some or all of the above processing in the testing unit may be performed using AI or not. For example, the testing unit can input user emotion data into a generative AI, which can then perform emotion estimation.

[0104] The testing department can determine the priority of the AI ​​agent tests based on the submission timing during testing. For example, the testing department can prioritize testing AI agents that submit earlier. Conversely, the testing department can postpone testing AI agents that submit later. Furthermore, the testing department can determine the priority of the tests based on the submission timing of the AI ​​agents. This allows for more efficient testing by prioritizing tests based on the submission timing of the AI ​​agents. Some or all of the above processes in the testing department may be performed using AI or not. For example, the testing department can input AI agent submission timing data into an AI, which can then determine the priority of the tests.

[0105] The testing unit can adjust the order of testing based on the relevance of the AI ​​agents during testing. For example, the testing unit can prioritize testing highly relevant AI agents. It can also postpone testing less relevant AI agents. Furthermore, the testing unit can adjust the order of testing based on the relevance of the AI ​​agents. This allows for more efficient testing by adjusting the order of testing based on the relevance of the AI ​​agents. Some or all of the above processing in the testing unit may be performed using AI or not. For example, the testing unit can input AI agent relevance data into an AI, which can then adjust the order of testing.

[0106] The review unit can estimate the user's emotions and adjust the review criteria based on the estimated emotions. For example, if the user is nervous, the review unit may relax the review criteria. Conversely, if the user is relaxed, the review unit may tighten the review criteria. Furthermore, if the user is in a hurry, the review unit may simplify the review criteria. This allows for more appropriate reviews by adjusting the review criteria according to the user's emotions. Emotion estimation is achieved using an emotion estimation function, such as an emotion engine or generative AI. Generative AI may be, but is not limited to, text generation AI (e.g., LLM) or multimodal generation AI. Some or all of the above processing in the review unit may be performed using AI or not. For example, the review unit can input user emotion data into a generative AI, which can then perform emotion estimation.

[0107] The review department can improve the accuracy of its reviews by considering the interrelationships of AI agents during the review process. For example, the review department can analyze the interrelationships of AI agents to improve the accuracy of the review. The review department can also adjust the review criteria by considering the interrelationships of AI agents. Furthermore, the review department can improve the accuracy of the review based on the interrelationships of AI agents. In this way, the accuracy of the review is improved by considering the interrelationships of AI agents. Some or all of the above processes in the review department may be performed using AI or not. For example, the review department can input data on the interrelationships of AI agents into an AI, which can then improve the accuracy of the review.

[0108] The review department can conduct its review by considering the attribute information of the submitter provided by the AI ​​agent. For example, the review department can adjust the review criteria based on the submitter's attribute information. The review department can also improve the accuracy of the review by considering the submitter's attribute information. Furthermore, the review department can conduct the review based on the submitter's attribute information. This improves the accuracy of the review by considering the submitter's attribute information. Some or all of the above processes in the review department may be performed using AI or not. For example, the review department can input the submitter's attribute information data into the AI, which can then improve the accuracy of the review.

[0109] The review unit can estimate the user's emotions and adjust the order in which the review results are displayed based on the estimated emotions. For example, if the user is nervous, the review unit can display important results first. If the user is relaxed, the review unit can also display detailed results sequentially. Furthermore, if the user is in a hurry, the review unit can display concise results first. By adjusting the display order of review results according to the user's emotions, it becomes possible to provide more appropriate review results. Emotion estimation is achieved using an emotion estimation function, such as an emotion engine or generative AI. Generative AI is, but is not limited to, text generation AI (e.g., LLM) or multimodal generation AI. Some or all of the above processing in the review unit may be performed using AI or not. For example, the review unit can input user emotion data into a generative AI, which can then perform emotion estimation.

[0110] The review department can conduct reviews while considering the geographical distribution of AI agents. For example, the review department can adjust the review criteria based on the geographical distribution of AI agents. The review department can also improve the accuracy of the review by considering the geographical distribution of AI agents. Furthermore, the review department can conduct reviews based on the geographical distribution of AI agents. This improves the accuracy of the review by considering the geographical distribution of AI agents. Some or all of the above processes in the review department may be performed using AI or not. For example, the review department can input geographical distribution data of AI agents into an AI, which can then improve the accuracy of the review.

[0111] The review department can improve the accuracy of its review by referring to the AI ​​agent's relevant literature during the review process. For example, the review department can improve the accuracy of its review by referring to the AI ​​agent's relevant literature. The review department can also adjust its review criteria based on the AI ​​agent's relevant literature. Furthermore, the review department can improve the accuracy of its review by considering the AI ​​agent's relevant literature. As a result, the accuracy of the review is improved by referring to the AI ​​agent's relevant literature. Some or all of the above processes in the review department may be performed using AI or not. For example, the review department can input the AI ​​agent's relevant literature data into the AI, which can then improve the accuracy of its review.

[0112] The recognition unit can estimate the user's emotions and adjust the recognition method based on the estimated emotions. For example, if the user is nervous, the recognition unit may relax the recognition method. Conversely, if the user is relaxed, the recognition unit may make the recognition method stricter. Furthermore, if the user is in a hurry, the recognition unit may simplify the recognition method. This allows for more appropriate recognition by adjusting the recognition method according to the user's emotions. Emotion estimation is achieved using an emotion estimation function, such as an emotion engine or generative AI. Generative AI may be, but is not limited to, text generation AI (e.g., LLM) or multimodal generation AI. Some or all of the above processing in the recognition unit may be performed using AI or not. For example, the recognition unit can input user emotion data into a generative AI, which can then perform emotion estimation.

[0113] The certification department can analyze the AI ​​agent's past activities during the certification process to select the optimal certification method. For example, the certification department can select the optimal certification method based on the AI ​​agent's past activity history. The certification department can also analyze the AI ​​agent's past activities and adjust the certification criteria. Furthermore, the certification department can customize the certification method considering the AI ​​agent's past activities. This allows the certification department to select the optimal certification method by analyzing the AI ​​agent's past activities. Some or all of the above processes in the certification department may be performed using AI or not. For example, the certification department can input the AI ​​agent's past activity data into an AI, which can then select the optimal certification method.

[0114] The certification unit can customize the certification process based on the current status of the AI ​​agent during the certification process. For example, the certification unit can customize the certification process based on the current status of the AI ​​agent. The certification unit can also adjust the certification criteria considering the current status of the AI ​​agent. Furthermore, the certification unit can select a certification method based on the current status of the AI ​​agent. This allows for more appropriate certification by customizing the certification process based on the current status of the AI ​​agent. Some or all of the above processes in the certification unit may be performed using AI or not. For example, the certification unit can input the current status data of the AI ​​agent into the AI, and the AI ​​can customize the certification process.

[0115] The recognition unit can estimate the user's emotions and determine recognition priorities based on the estimated emotions. For example, if the user is nervous, the recognition unit can lower the recognition priority. Conversely, if the user is relaxed, the recognition unit can raise the recognition priority. Furthermore, if the user is in a hurry, the recognition unit can also raise the recognition priority. This allows for more appropriate recognition by determining recognition priorities according to the user's emotions. Emotion estimation is achieved using an emotion estimation function, such as an emotion engine or generative AI. Generative AI may be, but is not limited to, text generation AI (e.g., LLM) or multimodal generation AI. Some or all of the above processing in the recognition unit may be performed using AI or not. For example, the recognition unit can input user emotion data into a generative AI, which can then perform emotion estimation.

[0116] The certification unit can select the optimal certification method by considering the geographical location information of the AI ​​agent during the certification process. For example, the certification unit can select the optimal certification method based on the geographical location information of the AI ​​agent. The certification unit can also adjust the certification criteria by considering the geographical location information of the AI ​​agent. Furthermore, the certification unit can customize the certification means based on the geographical location information of the AI ​​agent. This allows for the selection of the optimal certification method by considering the geographical location information of the AI ​​agent. Some or all of the above-described processes in the certification unit may be performed using AI or not. For example, the certification unit can input the geographical location data of the AI ​​agent into the AI, and the AI ​​can select the optimal certification method.

[0117] The certification department can analyze the social media activity of an AI agent during the certification process and propose certification methods. For example, the certification department can analyze the social media activity of an AI agent and propose the optimal certification method. The certification department can also adjust the certification criteria based on the social media activity of the AI ​​agent. Furthermore, the certification department can customize the certification methods considering the social media activity of the AI ​​agent. This allows the certification department to propose the optimal certification method by analyzing the social media activity of the AI ​​agent. Some or all of the above processes in the certification department may be performed using AI or not. For example, the certification department can input the social media activity data of the AI ​​agent into an AI, which can then propose the optimal certification method.

[0118] The audit department can estimate the user's emotions and adjust the audit method based on the estimated emotions. For example, if the user is tense, the audit department may relax the audit method. Conversely, if the user is relaxed, the audit department may make the audit method more rigorous. Furthermore, if the user is in a hurry, the audit department may simplify the audit method. This allows for more appropriate audits by adjusting the audit method according to the user's emotions. Emotion estimation is achieved using an emotion estimation function, such as an emotion engine or generative AI. Generative AI may be, but is not limited to, text generation AI (e.g., LLM) or multimodal generation AI. Some or all of the above processing in the audit department may be performed using AI or not. For example, the audit department can input user emotion data into a generative AI, which can then perform emotion estimation.

[0119] The audit department can analyze the AI ​​agent's past activities during an audit to select the optimal audit method. For example, the audit department can select the optimal audit method based on the AI ​​agent's past activity history. The audit department can also analyze the AI ​​agent's past activities and adjust the audit standards. Furthermore, the audit department can customize the audit method considering the AI ​​agent's past activities. This allows the audit department to select the optimal audit method by analyzing the AI ​​agent's past activities. Some or all of the above processes in the audit department may be performed using AI or not. For example, the audit department can input the AI ​​agent's past activity data into the AI, which can then select the optimal audit method.

[0120] The audit department can customize audit methods based on the current status of the AI ​​agent during an audit. For example, the audit department can customize audit methods based on the current status of the AI ​​agent. The audit department can also adjust audit criteria considering the current status of the AI ​​agent. Furthermore, the audit department can select audit methods based on the current status of the AI ​​agent. This allows for more appropriate audits by customizing audit methods based on the current status of the AI ​​agent. Some or all of the above processes in the audit department may be performed using AI or not. For example, the audit department can input current status data of the AI ​​agent into the AI, and the AI ​​can customize the audit methods.

[0121] The audit department can estimate the user's emotions and determine audit priorities based on those estimated emotions. For example, if the user is stressed, the audit department may lower the audit priority. Conversely, if the user is relaxed, the audit department may raise the audit priority. Furthermore, if the user is in a hurry, the audit department may raise the audit priority. This allows for more appropriate audits by determining audit priorities according to the user's emotions. Emotion estimation is achieved using an emotion estimation function, such as an emotion engine or generative AI. Generative AI may be, but is not limited to, text generation AI (e.g., LLM) or multimodal generation AI. Some or all of the above processing in the audit department may be performed using AI or not. For example, the audit department can input user emotion data into a generative AI, which can then estimate the emotions.

[0122] The audit department can select the optimal audit method during an audit by considering the geographical location information of the AI ​​agent. For example, the audit department can select the optimal audit method based on the geographical location information of the AI ​​agent. The audit department can also adjust the audit criteria by considering the geographical location information of the AI ​​agent. Furthermore, the audit department can customize the audit methods based on the geographical location information of the AI ​​agent. This allows for the selection of the optimal audit method by considering the geographical location information of the AI ​​agent. Some or all of the above processes performed by the audit department may be carried out using AI or not. For example, the audit department can input the geographical location data of the AI ​​agent into the AI, and the AI ​​can select the optimal audit method.

[0123] The audit department can analyze the social media activity of AI agents during an audit and propose audit methods. For example, the audit department can analyze the social media activity of AI agents and propose the most suitable audit methods. The audit department can also adjust audit standards based on the social media activity of AI agents. Furthermore, the audit department can customize audit methods considering the social media activity of AI agents. This allows the audit department to propose the most suitable audit methods by analyzing the social media activity of AI agents. Some or all of the above processes performed by the audit department may be carried out using AI or not. For example, the audit department can input the social media activity data of AI agents into an AI, which can then propose the most suitable audit methods.

[0124] The robot instruction unit can estimate the user's emotions and adjust the way it instructs the robot based on the estimated emotions. For example, if the user is nervous, the robot instruction unit can provide simple and easy-to-understand instructions. If the user is relaxed, it can also provide instructions that include detailed information. Furthermore, if the user is in a hurry, it can provide concise instructions. By adjusting the way it instructs the robot according to the user's emotions, more appropriate instructions can be provided. Emotion estimation is achieved using an emotion estimation function, such as an emotion engine or a generative AI. The generative AI is, but is not limited to, text generation AI (e.g., LLM) or multimodal generation AI. Some or all of the above processing in the robot instruction unit may be performed using AI or not. For example, the robot instruction unit can input user emotion data into a generative AI, which can then perform emotion estimation.

[0125] The robot instruction unit can analyze the AI ​​agent's past instruction history to select the optimal instruction method when giving instructions to the robot. For example, the robot instruction unit can select the optimal instruction method based on the AI ​​agent's past instruction history. The robot instruction unit can also analyze the AI ​​agent's past instruction history and adjust the instruction criteria. Furthermore, the robot instruction unit can customize the instruction method considering the AI ​​agent's past instruction history. This allows the optimal instruction method to be selected by analyzing the AI ​​agent's past instruction history. Some or all of the above processes in the robot instruction unit may be performed using AI or not. For example, the robot instruction unit can input the AI ​​agent's past instruction history data into the AI, and the AI ​​can select the optimal instruction method.

[0126] The robot instruction unit can customize the instruction methods based on the current status of the AI ​​agent when giving instructions to the robot. For example, the robot instruction unit customizes the instruction methods based on the current status of the AI ​​agent. The robot instruction unit can also adjust the instruction criteria considering the current status of the AI ​​agent. Furthermore, the robot instruction unit can select an instruction method based on the current status of the AI ​​agent. This allows for more appropriate instructions by customizing the instruction methods based on the current status of the AI ​​agent. Some or all of the above processes in the robot instruction unit may be performed using AI or not. For example, the robot instruction unit can input the current status data of the AI ​​agent into the AI, and the AI ​​can customize the instruction methods.

[0127] The robot command unit can estimate the user's emotions and determine the priority of instructions to the robot based on the estimated emotions. For example, if the user is nervous, the robot command unit can lower the priority of instructions. Conversely, if the user is relaxed, the robot command unit can also raise the priority of instructions. Furthermore, if the user is in a hurry, the robot command unit can also raise the priority of instructions. This allows for more appropriate instructions by determining the priority of instructions to the robot according to the user's emotions. Emotion estimation is achieved using an emotion estimation function, such as an emotion engine or a generative AI. The generative AI is, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI. Some or all of the above processing in the robot command unit may be performed using AI or not. For example, the robot command unit can input user emotion data into a generative AI, which can then perform emotion estimation.

[0128] The robot instruction unit can select the optimal instruction method when giving instructions to the robot, taking into account the geographical location information of the AI ​​agent. For example, the robot instruction unit can select the optimal instruction method based on the geographical location information of the AI ​​agent. The robot instruction unit can also adjust the instruction criteria, taking into account the geographical location information of the AI ​​agent. Furthermore, the robot instruction unit can customize the instruction means based on the geographical location information of the AI ​​agent. This allows for the selection of the optimal instruction method by considering the geographical location information of the AI ​​agent. Some or all of the above processing in the robot instruction unit may be performed using AI, or not using AI. For example, the robot instruction unit can input the geographical location data of the AI ​​agent into the AI, and the AI ​​can select the optimal instruction method.

[0129] The robot instruction unit can analyze the AI ​​agent's social media activity and propose instruction methods when issuing robot instructions. For example, the robot instruction unit can analyze the AI ​​agent's social media activity and propose the optimal instruction method. The robot instruction unit can also adjust instruction criteria based on the AI ​​agent's social media activity. Furthermore, the robot instruction unit can customize instruction methods considering the AI ​​agent's social media activity. This allows the optimal instruction method to be proposed by analyzing the AI ​​agent's social media activity. Some or all of the above processing in the robot instruction unit may be performed using AI or not. For example, the robot instruction unit can input the AI ​​agent's social media activity data into the AI, which can then propose the optimal instruction method.

[0130] The robot evaluation unit can estimate the user's emotions and adjust the robot's evaluation method based on the estimated user emotions. For example, if the user is nervous, the robot evaluation unit can provide a simple and highly visual evaluation method. If the user is relaxed, the robot evaluation unit can also provide an evaluation method that includes detailed information. Furthermore, if the user is in a hurry, the robot evaluation unit can provide a concise evaluation method. By adjusting the robot's evaluation method according to the user's emotions, a more appropriate evaluation becomes possible. Emotion estimation is achieved using an emotion estimation function, such as an emotion engine or a generative AI. The generative AI is, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI. Some or all of the above processing in the robot evaluation unit may be performed using AI or not. For example, the robot evaluation unit can input user emotion data into a generative AI, which can then perform emotion estimation.

[0131] The robot evaluation unit can select the optimal evaluation method by analyzing the past evaluation history of the AI ​​agent during robot evaluation. For example, the robot evaluation unit selects the optimal evaluation method based on the past evaluation history of the AI ​​agent. The robot evaluation unit can also analyze the past evaluation history of the AI ​​agent and adjust the evaluation criteria. Furthermore, the robot evaluation unit can customize the evaluation method by considering the past evaluation history of the AI ​​agent. This allows the robot evaluation unit to select the optimal evaluation method by analyzing the past evaluation history of the AI ​​agent. Some or all of the above processes in the robot evaluation unit may be performed using AI or not. For example, the robot evaluation unit can input the past evaluation history data of the AI ​​agent into the AI, and the AI ​​can select the optimal evaluation method.

[0132] The robot evaluation unit can customize the evaluation methods based on the current status of the AI ​​agent during robot evaluation. For example, the robot evaluation unit customizes the evaluation methods based on the current status of the AI ​​agent. The robot evaluation unit can also adjust the evaluation criteria considering the current status of the AI ​​agent. Furthermore, the robot evaluation unit can select an evaluation method based on the current status of the AI ​​agent. This allows for a more appropriate evaluation by customizing the evaluation methods based on the current status of the AI ​​agent. Some or all of the above processes in the robot evaluation unit may be performed using AI or not. For example, the robot evaluation unit can input the current status data of the AI ​​agent into the AI, and the AI ​​can customize the evaluation methods.

[0133] The robot evaluation unit can estimate the user's emotions and determine the priority of the robot evaluation based on the estimated user emotions. For example, if the user is nervous, the robot evaluation unit may lower the evaluation priority. Conversely, if the user is relaxed, the robot evaluation unit may raise the evaluation priority. Furthermore, if the user is in a hurry, the robot evaluation unit may also raise the evaluation priority. This allows for more appropriate evaluation by determining the priority of the robot evaluation according to the user's emotions. Emotion estimation is achieved using an emotion estimation function, such as an emotion engine or a generative AI. The generative AI is, but is not limited to, text generation AI (e.g., LLM) or multimodal generation AI. Some or all of the above processing in the robot evaluation unit may be performed using AI or not. For example, the robot evaluation unit can input user emotion data into a generative AI, which can then perform emotion estimation.

[0134] The robot evaluation unit can select the optimal evaluation method when evaluating a robot, taking into account the geographical location information of the AI ​​agent. For example, the robot evaluation unit can select the optimal evaluation method based on the geographical location information of the AI ​​agent. The robot evaluation unit can also adjust the evaluation criteria, taking into account the geographical location information of the AI ​​agent. Furthermore, the robot evaluation unit can customize the evaluation means based on the geographical location information of the AI ​​agent. This allows for the selection of the optimal evaluation method by considering the geographical location information of the AI ​​agent. Some or all of the above-described processes in the robot evaluation unit may be performed using AI, or they may not be performed using AI. For example, the robot evaluation unit can input the geographical location data of the AI ​​agent into the AI, and the AI ​​can select the optimal evaluation method.

[0135] The robot evaluation unit can analyze the social media activity of an AI agent during robot evaluation and propose evaluation methods. For example, the robot evaluation unit can analyze the social media activity of an AI agent and propose the optimal evaluation method. The robot evaluation unit can also adjust the evaluation criteria based on the social media activity of the AI ​​agent. Furthermore, the robot evaluation unit can customize the evaluation method by taking into account the social media activity of the AI ​​agent. This allows the robot evaluation unit to propose the optimal evaluation method by analyzing the social media activity of the AI ​​agent. Some or all of the above processing in the robot evaluation unit may be performed using AI or not. For example, the robot evaluation unit can input the social media activity data of the AI ​​agent into the AI, and the AI ​​can propose the optimal evaluation method.

[0136] The dialogue evaluation unit can estimate the user's emotions and adjust the dialogue evaluation method based on the estimated user emotions. For example, if the user is nervous, the dialogue evaluation unit can provide a simple and highly visual dialogue evaluation method. If the user is relaxed, the dialogue evaluation unit can also provide a dialogue evaluation method that includes detailed information. Furthermore, if the user is in a hurry, the dialogue evaluation unit can provide a dialogue evaluation method that gets straight to the point. By adjusting the dialogue evaluation method according to the user's emotions, a more appropriate evaluation becomes possible. Emotion estimation is achieved using an emotion estimation function, such as an emotion engine or a generative AI. The generative AI is, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI. Some or all of the above processing in the dialogue evaluation unit may be performed using the generative AI or not. For example, the dialogue evaluation unit can input user emotion data into the generative AI, and the generative AI can perform emotion estimation.

[0137] The dialogue evaluation unit can analyze the AI ​​agent's past dialogue history during dialogue evaluation to select the optimal evaluation method. For example, the dialogue evaluation unit selects the optimal evaluation method based on the AI ​​agent's past dialogue history. The dialogue evaluation unit can also analyze the AI ​​agent's past dialogue history and adjust the evaluation criteria. Furthermore, the dialogue evaluation unit can customize the evaluation method by considering the AI ​​agent's past dialogue history. This allows the optimal evaluation method to be selected by analyzing the AI ​​agent's past dialogue history. Some or all of the above processing in the dialogue evaluation unit may be performed using a generative AI, or it may be performed without a generative AI. For example, the dialogue evaluation unit can input the AI ​​agent's past dialogue history data into a generative AI, which can then select the optimal evaluation method.

[0138] The dialogue evaluation unit can customize the evaluation methods based on the current state of the AI ​​agent during dialogue evaluation. For example, the dialogue evaluation unit customizes the evaluation methods based on the current state of the AI ​​agent. The dialogue evaluation unit can also adjust the evaluation criteria considering the current state of the AI ​​agent. Furthermore, the dialogue evaluation unit can select an evaluation method based on the current state of the AI ​​agent. This allows for a more appropriate evaluation by customizing the evaluation methods based on the current state of the AI ​​agent. Some or all of the above processing in the dialogue evaluation unit may be performed using a generating AI, or it may be performed without a generating AI. For example, the dialogue evaluation unit can input the current state data of the AI ​​agent into a generating AI, and the generating AI can customize the evaluation methods.

[0139] The dialogue evaluation unit can estimate the user's emotions and determine the priority of the dialogue evaluation based on the estimated user emotions. For example, if the user is nervous, the dialogue evaluation unit may lower the evaluation priority. Conversely, if the user is relaxed, the dialogue evaluation unit may raise the evaluation priority. Furthermore, if the user is in a hurry, the dialogue evaluation unit may also raise the evaluation priority. This allows for more appropriate evaluation by determining the priority of the dialogue evaluation according to the user's emotions. Emotion estimation is achieved using an emotion estimation function, such as an emotion engine or a generative AI. The generative AI is, but is not limited to, text generation AI (e.g., LLM) or multimodal generation AI. Some or all of the above processing in the dialogue evaluation unit may be performed using a generative AI or not. For example, the dialogue evaluation unit can input user emotion data into a generative AI, which can then perform emotion estimation.

[0140] The dialogue evaluation unit can select the optimal evaluation method during dialogue evaluation by considering the geographical location information of the AI ​​agent. For example, the dialogue evaluation unit selects the optimal evaluation method based on the geographical location information of the AI ​​agent. The dialogue evaluation unit can also adjust the evaluation criteria by considering the geographical location information of the AI ​​agent. Furthermore, the dialogue evaluation unit can customize the evaluation means based on the geographical location information of the AI ​​agent. This allows for the selection of the optimal evaluation method by considering the geographical location information of the AI ​​agent. Some or all of the above processing in the dialogue evaluation unit may be performed using a generating AI, or it may be performed without using a generating AI. For example, the dialogue evaluation unit can input the geographical location information data of the AI ​​agent into a generating AI, and the generating AI can select the optimal evaluation method.

[0141] The dialogue evaluation unit can analyze the AI ​​agent's social media activity during dialogue evaluation and propose evaluation methods. For example, the dialogue evaluation unit can analyze the AI ​​agent's social media activity and propose the optimal evaluation method. The dialogue evaluation unit can also adjust the evaluation criteria based on the AI ​​agent's social media activity. Furthermore, the dialogue evaluation unit can customize the evaluation method by taking the AI ​​agent's social media activity into consideration. This allows the optimal evaluation method to be proposed by analyzing the AI ​​agent's social media activity. Some or all of the above processing in the dialogue evaluation unit may be performed using a generative AI, or it may be performed without a generative AI. For example, the dialogue evaluation unit can input the AI ​​agent's social media activity data into a generative AI, and the generative AI can propose the optimal evaluation method.

[0142] The instruction analysis unit can estimate the user's emotions and adjust the instruction analysis method based on the estimated user emotions. For example, if the user is nervous, the instruction analysis unit can provide a simple and highly visible instruction analysis method. If the user is relaxed, the instruction analysis unit can also provide an instruction analysis method that includes detailed information. Furthermore, if the user is in a hurry, the instruction analysis unit can provide an instruction analysis method that gets straight to the point. By adjusting the instruction analysis method according to the user's emotions, more appropriate analysis becomes possible. Emotion estimation is achieved using an emotion estimation function, such as an emotion engine or a generative AI. The generative AI is, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI. Some or all of the above-described processing in the instruction analysis unit may be performed using the generative AI or not. For example, the instruction analysis unit can input user emotion data into the generative AI, and the generative AI can perform emotion estimation.

[0143] The instruction analysis unit can analyze the AI ​​agent's past instruction history and select the optimal analysis method during instruction analysis. For example, the instruction analysis unit selects the optimal analysis method based on the AI ​​agent's past instruction history. The instruction analysis unit can also analyze the AI ​​agent's past instruction history and adjust the analysis criteria. Furthermore, the instruction analysis unit can customize the analysis method considering the AI ​​agent's past instruction history. This allows the optimal analysis method to be selected by analyzing the AI ​​agent's past instruction history. Some or all of the above processing in the instruction analysis unit may be performed using a generating AI, or it may be performed without using a generating AI. For example, the instruction analysis unit can input the AI ​​agent's past instruction history data into a generating AI, which can then select the optimal analysis method.

[0144] The instruction analysis unit can customize the analysis methods based on the current status of the AI ​​agent during instruction analysis. For example, the instruction analysis unit customizes the analysis methods based on the current status of the AI ​​agent. The instruction analysis unit can also adjust the analysis criteria considering the current status of the AI ​​agent. Furthermore, the instruction analysis unit can select an analysis method based on the current status of the AI ​​agent. This allows for more appropriate analysis by customizing the analysis methods based on the current status of the AI ​​agent. Some or all of the above processing in the instruction analysis unit may be performed using a generating AI, or it may be performed without using a generating AI. For example, the instruction analysis unit can input the current status data of the AI ​​agent into a generating AI, and the generating AI can customize the analysis methods.

[0145] The instruction analysis unit can estimate the user's emotions and determine the priority of instruction analysis based on the estimated user emotions. For example, if the user is tense, the instruction analysis unit can lower the priority of the analysis. Conversely, if the user is relaxed, the instruction analysis unit can also raise the priority of the analysis. Furthermore, if the user is in a hurry, the instruction analysis unit can also raise the priority of the analysis. This allows for more appropriate analysis by determining the priority of instruction analysis according to the user's emotions. Emotion estimation is achieved using an emotion estimation function, such as an emotion engine or a generative AI. The generative AI is, but is not limited to, a text generation AI (e.g., LLM) or a multimodal generation AI. Some or all of the above-described processing in the instruction analysis unit may be performed using the generative AI or not. For example, the instruction analysis unit can input user emotion data into the generative AI, which can then perform emotion estimation.

[0146] The instruction analysis unit can select the optimal analysis method by considering the geographical location information of the AI ​​agent during instruction analysis. For example, the instruction analysis unit selects the optimal analysis method based on the geographical location information of the AI ​​agent. The instruction analysis unit can also adjust the analysis criteria by considering the geographical location information of the AI ​​agent. Furthermore, the instruction analysis unit can customize the analysis means based on the geographical location information of the AI ​​agent. This allows for the selection of the optimal analysis method by considering the geographical location information of the AI ​​agent. Some or all of the above-described processes in the instruction analysis unit may be performed using a generating AI, or they may be performed without using a generating AI. For example, the instruction analysis unit can input the geographical location information data of the AI ​​agent into a generating AI, and the generating AI can select the optimal analysis method.

[0147] The instruction analysis unit can analyze the AI ​​agent's social media activities during instruction analysis and propose analysis methods. For example, the instruction analysis unit can analyze the AI ​​agent's social media activities and propose the optimal analysis method. The instruction analysis unit can also adjust the analysis criteria based on the AI ​​agent's social media activities. Furthermore, the instruction analysis unit can customize the analysis method considering the AI ​​agent's social media activities. This allows the optimal analysis method to be proposed by analyzing the AI ​​agent's social media activities. Some or all of the above processing in the instruction analysis unit may be performed using a generating AI, or it may be performed without using a generating AI. For example, the instruction analysis unit can input the AI ​​agent's social media activity data into a generating AI, and the generating AI can propose the optimal analysis method.

[0148] The system according to the embodiment is not limited to the example described above, and various modifications are possible, for example, as follows.

[0149] The AI ​​agent certification system allows the application processing department to analyze an applicant's past application history and propose the optimal application processing method. For example, applicants who have previously requested expedited processing will have their applications prioritized. Furthermore, applicants who have submitted complex applications in the past can be provided with detailed guidelines. The system can also analyze applicant trends based on their past application history and customize the optimal application processing method. This allows for more appropriate application processing by considering the applicant's past application history.

[0150] The AI ​​agent certification system allows the examination department to analyze the learning history of AI agents and provide optimal exam content. For example, an AI agent that has achieved high scores in a particular field in the past will be given an advanced exam related to that field. Similarly, an AI agent that has struggled in a particular area in the past will be given an exam focused on that area. Furthermore, the exam content can be customized based on the learning history to promote the AI ​​agent's growth. This allows for the provision of more appropriate exams by considering the AI ​​agent's learning history.

[0151] The AI ​​agent certification system allows the review department to implement peer evaluation of AI agents, thereby improving the accuracy of the review process. For example, it can collect feedback from other AI agents and use it as a reference for the review. Furthermore, based on peer evaluation, it can analyze the strengths and weaknesses of AI agents and adjust the review criteria. It can also utilize peer evaluation to assess the collaborative relationships between AI agents. In this way, implementing peer evaluation of AI agents improves the accuracy of the review process.

[0152] The AI ​​agent certification system allows the certification department to publish the performance of AI agents, thereby increasing transparency. For example, the performance of certified AI agents can be made public online, allowing other applicants and stakeholders to view it. Furthermore, the certification criteria can be reviewed based on the performance, and stricter standards can be set. In addition, the reliability of AI agents can be improved through the publication of performance. In this way, by disclosing the performance of AI agents, the transparency of the certification process is increased and the reliability is improved.

[0153] The AI ​​agent certification system allows the audit department to monitor AI agent activities in real time, enabling rapid response. For example, it can monitor AI agent activity logs in real time and respond immediately if an anomaly occurs. Furthermore, based on real-time monitoring, the AI ​​agent's performance can be evaluated, and improvement guidance can be provided as needed. In addition, real-time monitoring can increase the transparency of AI agent activities. This results in faster response times and improved transparency through real-time monitoring of AI agent activities.

[0154] The application processing unit can estimate the user's emotions and customize the application processing method based on those emotions. For example, if the user is nervous, it can provide a simple and highly visible application form. If the user is relaxed, it can provide an application form with more detailed information. Furthermore, if the user is in a hurry, it can provide a concise application form. By customizing the application processing method according to the user's emotions, more appropriate application processing becomes possible.

[0155] The testing department can estimate the user's emotions and adjust the difficulty of the test based on those estimates. For example, if the user is nervous, it can provide a test with a lower difficulty level. Conversely, if the user is relaxed, it can provide a test with a higher difficulty level. Furthermore, if the user is in a hurry, it can provide a test that can be completed in a short amount of time. In this way, by adjusting the difficulty of the test according to the user's emotions, a more appropriate test can be provided.

[0156] The review department can estimate the user's emotions and adjust the review feedback based on those estimates. For example, if the user is nervous, positive feedback will be prioritized. If the user is relaxed, detailed feedback may be provided. Furthermore, if the user is in a hurry, concise feedback may be provided. By adjusting the review feedback according to the user's emotions, more appropriate feedback can be provided.

[0157] The certification unit can estimate the user's emotions and adjust the certification notification method based on those emotions. For example, if the user is stressed, it can provide a simple, highly visible notification. If the user is relaxed, it can provide a notification with more detailed information. Furthermore, if the user is in a hurry, it can provide a concise notification. By adjusting the certification notification method according to the user's emotions, more appropriate notifications can be provided.

[0158] The audit department can estimate the user's emotions and adjust the audit reporting method based on that estimation. For example, if the user is stressed, a simple and easy-to-read report can be provided. If the user is relaxed, a report with more detailed information can be provided. Furthermore, if the user is in a hurry, a concise report can be provided. By adjusting the audit reporting method according to the user's emotions, more appropriate reporting becomes possible.

[0159] The following briefly describes the processing flow for example form 2.

[0160] Step 1: The application receiving department accepts the application. The application receiving department can accept applications in various formats, such as paper applications and online applications. Step 2: The Testing Department tests the applications received by the Application Acceptance Department. The Testing Department may, for example, test the skills and ethical aspects of the AI ​​agent. The Testing Department may conduct tests to evaluate the AI ​​agent's programming skills and problem-solving abilities. The Testing Department may also conduct tests to evaluate the AI ​​agent's ethical judgment abilities. Step 3: The Review Department reviews the results of the tests conducted by the Test Department. The Review Department evaluates, for example, the accuracy and reliability of the test results. Based on the test results, the Review Department decides whether or not to certify the AI ​​agent. Step 4: The Certification Department certifies the AI ​​agents based on the results of the review conducted by the Examination Department. The Certification Department certifies, for example, whether the AI ​​agents' skills and ethical aspects meet the standards. The Certification Department can issue a certificate to the certified AI agents. Step 5: The Audit Department audits the activities of AI agents certified by the Certification Department. For example, the Audit Department periodically audits whether the AI ​​agents' activities meet the certification standards. The Audit Department can monitor the AI ​​agents' activities and provide guidance for improvement as needed.

[0161] 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 user input for the result of the specific processing. The control unit 46A transmits the audio data indicating 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.

[0162] Data generation model 58 is a form of so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> Examples of generative AI include text generation AI, image generation AI, and multimodal generation AI. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images (e.g., still image data or video data). The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference result in one or more data formats from audio data, text data, and image data. The data generation model 58 includes, for example, text generation AI, image generation AI, and multimodal generation AI. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The specific processing unit 290 performs the specific processing described above using the data generation model 58. The data generation model 58 may be a fine-tuned model that outputs inference results from prompts that do not contain instructions, in which case the data generation model 58 can output inference results from prompts that do not contain instructions. In the data processing device 12, etc., there are multiple types of data generation models 58, and the data generation model 58 includes AI other than generative AI. AI other than generative AI includes, for example, linear regression, logistic regression, decision trees, random forests, support vector machines (SVMs), k-means clustering, convolutional neural networks (CNNs), recurrent neural networks (RNNs), generative adversarial networks (GANs), or naive Bayes, and can perform various processes, but is not limited to these examples. Also, the AI ​​may be an AI agent. Furthermore, when the processing of each of the above parts is performed by the AI, the processing may be performed by the AI ​​in part or in whole, but is not limited to this example.Furthermore, processing performed by AI, including generative AI, may be replaced with rule-based processing, and rule-based processing may be replaced with processing performed by AI, including generative AI.

[0163] Furthermore, the processing performed by the data processing system 10 described above is carried out by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the smart device 14, but it may also be carried out by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the smart device 14. In addition, the specific processing unit 290 of the data processing device 12 acquires or collects information necessary for processing from the smart device 14 or an external device, and the smart device 14 acquires or collects information necessary for processing from the data processing device 12 or an external device.

[0164] Each of the multiple elements described above, including the application acceptance unit, testing unit, review unit, certification unit, audit unit, robot instruction unit, robot evaluation unit, dialogue evaluation unit, and instruction analysis unit, is implemented by at least one of the smart device 14 and the data processing unit 12. For example, the application acceptance unit is implemented by the control unit 46A of the smart device 14 and can accept online applications. The testing unit is implemented by the specific processing unit 290 of the data processing unit 12 and tests the skills and ethical aspects of the AI ​​agent. The review unit is implemented by the specific processing unit 290 of the data processing unit 12 and reviews the test results. The certification unit is implemented by the specific processing unit 290 of the data processing unit 12 and performs certification. The audit unit is implemented by the specific processing unit 290 of the data processing unit 12 and audits the activities of the certified AI agent. The robot instruction unit is implemented by the control unit 46A of the smart device 14 and the AI ​​agent issues instructions to the physical robot. The robot evaluation unit is implemented by the specific processing unit 290 of the data processing unit 12 and evaluates the robot's operation. The dialogue evaluation unit is implemented by the specific processing unit 290 of the data processing device 12 and evaluates the dialogue capabilities of the AI ​​agent using generated AI. The instruction analysis unit is implemented by the specific processing unit 290 of the data processing device 12 and analyzes the instructions of the AI ​​agent using generated AI and issues appropriate operation instructions to the physical robot. The correspondence between each unit and the device and control unit is not limited to the example described above and can be modified in various ways.

[0165] [Second Embodiment] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0166] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0167] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN and / or LAN.

[0168] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0169] The microphone 238 receives voice commands and other instructions from the user by receiving voice signals. The microphone 238 captures the voice signals from the user, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0170] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, which captures images of the area around the user (for example, an imaging range defined by a field of view equivalent to the field of vision of a typical healthy person).

[0171] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0172] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing by the processor 28. The storage 32 stores the specific processing program 56.

[0173] The processor 28 reads a 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 acting as a specific processing unit 290 according to the specific processing program 56 executed on the RAM 30.

[0174] Storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290. The identification processing unit 290 can estimate the user's emotions using the emotion identification model 59 and perform identification processing using the user's emotions. The emotion estimation function (emotion identification function) using the emotion identification model 59 performs various estimations and predictions regarding the user's emotions, including but not limited to these examples. Furthermore, emotion estimation and prediction also include, for example, emotion analysis.

[0175] In the smart glasses 214, specific processing is performed by the processor 46. The storage 50 stores a specific processing program 60. The processor 46 reads the specific processing program 60 from the storage 50 and executes the read specific processing program 60 on the RAM 48. The specific processing is realized by the processor 46 acting as a control unit 46A according to the specific processing program 60 executed on the RAM 48. The smart glasses 214 also have a data generation model 58 and an emotion identification model 59, similar to the data generation model and emotion identification model 59, and can perform processing similar to that of the specific processing unit 290 using these models.

[0176] Furthermore, other devices besides the data processing device 12 may also have the data generation model 58. For example, a server device may have the data generation model 58. In this case, the data processing device 12 obtains processing results (such as prediction results) using the data generation model 58 by communicating with the server device that has the data generation model 58. Also, the data processing device 12 may be a server device or a terminal device owned by the user (for example, a mobile phone, robot, home appliance, etc.).

[0177] 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 user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0178] The data generation model 58 is a so-called generative AI. An example of a data generation model 58 is a generative AI such as ChatGPT. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and inference data such as audio data representing speech, text data representing text, and image data representing images (e.g., still image data or video data). The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference result in one or more data formats such as audio data, text data, and image data. The data generation model 58 includes, for example, text generation AI, image generation AI, and multimodal generation AI. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The specific processing unit 290 performs the specific processing described above using the data generation model 58. The data generation model 58 may be a fine-tuned model that outputs inference results from prompts that do not contain instructions, in which case the data generation model 58 can output inference results from prompts that do not contain instructions. In the data processing device 12, etc., there are multiple types of data generation models 58, and the data generation model 58 includes AI other than generative AI. AI other than generative AI includes, for example, linear regression, logistic regression, decision trees, random forests, support vector machines (SVM), k-means clustering, convolutional neural networks (CNN), recurrent neural networks (RNN), generative adversarial networks (GAN), or naive Bayes, and can perform various processes, but is not limited to these examples. Also, the AI ​​may be an AI agent. Furthermore, when the processing of each part described above is performed by the AI, the processing may be performed by the AI ​​in part or in whole, but is not limited to this example. Also, processing performed by an AI including a generative AI may be replaced by rule-based processing, and rule-based processing may be replaced by processing performed by an AI including a generative AI.

[0179] The data processing system 210 according to the second embodiment performs the same processing as the data processing system 10 according to the first embodiment. The processing by the data processing system 210 is performed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the smart glasses 214, but it may also be performed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the smart glasses 214. In addition, the specific processing unit 290 of the data processing device 12 acquires or collects information necessary for processing from the smart glasses 214 or an external device, and the smart glasses 214 acquires or collects information necessary for processing from the data processing device 12 or an external device.

[0180] Each of the multiple elements described above, including the application acceptance unit, testing unit, review unit, certification unit, audit unit, robot instruction unit, robot evaluation unit, dialogue evaluation unit, and instruction analysis unit, is implemented by at least one of the smart glasses 214 and the data processing unit 12. For example, the application acceptance unit is implemented by the control unit 46A of the smart glasses 214 and can accept online applications. The testing unit is implemented by the specific processing unit 290 of the data processing unit 12 and tests the skills and ethical aspects of the AI ​​agent. The review unit is implemented by the specific processing unit 290 of the data processing unit 12 and reviews the test results. The certification unit is implemented by the specific processing unit 290 of the data processing unit 12 and performs certification. The audit unit is implemented by the specific processing unit 290 of the data processing unit 12 and audits the activities of the certified AI agent. The robot instruction unit is implemented by the control unit 46A of the smart glasses 214 and the AI ​​agent issues instructions to the physical robot. The robot evaluation unit is implemented by the specific processing unit 290 of the data processing unit 12 and evaluates the robot's operation. The dialogue evaluation unit is implemented by the specific processing unit 290 of the data processing device 12 and evaluates the dialogue capabilities of the AI ​​agent using generated AI. The instruction analysis unit is implemented by the specific processing unit 290 of the data processing device 12 and analyzes the instructions of the AI ​​agent using generated AI and issues appropriate operation instructions to the physical robot. The correspondence between each unit and the device and control unit is not limited to the example described above and can be modified in various ways.

[0181] [Third Embodiment] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0182] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0183] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN and / or LAN.

[0184] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[0185] The microphone 238 receives voice commands and other instructions from the user by receiving voice signals. The microphone 238 captures the voice signals from the user, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0186] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, which captures images of the area around the user (for example, an imaging range defined by a field of view equivalent to the field of vision of a typical healthy person).

[0187] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0188] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0189] The processor 28 reads a 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 acting as a specific processing unit 290 according to the specific processing program 56 executed on the RAM 30.

[0190] Storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290. The identification processing unit 290 can estimate the user's emotions using the emotion identification model 59 and perform identification processing using the user's emotions. The emotion estimation function (emotion identification function) using the emotion identification model 59 performs various estimations and predictions regarding the user's emotions, including but not limited to these examples. Furthermore, emotion estimation and prediction also include, for example, emotion analysis.

[0191] In the headset terminal 314, specific processing is performed by the processor 46. The storage 50 stores a specific program 60. The processor 46 reads the specific program 60 from the storage 50 and executes the read specific program 60 on the RAM 48. The specific processing is realized by the processor 46 acting as a control unit 46A according to the specific program 60 executed on the RAM 48. The headset terminal 314 also has a data generation model 58 and an emotion identification model 59, similar to the data generation model and emotion identification model 59, and can perform processing similar to that of the specific processing unit 290 using these models.

[0192] Furthermore, other devices besides the data processing device 12 may also have the data generation model 58. For example, a server device may have the data generation model 58. In this case, the data processing device 12 obtains processing results (such as prediction results) using the data generation model 58 by communicating with the server device that has the data generation model 58. Also, the data processing device 12 may be a server device or a terminal device owned by the user (for example, a mobile phone, robot, home appliance, etc.).

[0193] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0194] The data generation model 58 is a so-called generative AI. An example of a data generation model 58 is a generative AI such as ChatGPT. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and inference data such as audio data representing speech, text data representing text, and image data representing images (e.g., still image data or video data). The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference result in one or more data formats such as audio data, text data, and image data. The data generation model 58 includes, for example, text generation AI, image generation AI, and multimodal generation AI. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The specific processing unit 290 performs the specific processing described above using the data generation model 58. The data generation model 58 may be a fine-tuned model that outputs inference results from prompts that do not contain instructions, in which case the data generation model 58 can output inference results from prompts that do not contain instructions. In the data processing device 12, etc., there are multiple types of data generation models 58, and the data generation model 58 includes AI other than generative AI. AI other than generative AI includes, for example, linear regression, logistic regression, decision trees, random forests, support vector machines (SVM), k-means clustering, convolutional neural networks (CNN), recurrent neural networks (RNN), generative adversarial networks (GAN), or naive Bayes, and can perform various processes, but is not limited to these examples. Also, the AI ​​may be an AI agent. Furthermore, when the processing of each part described above is performed by the AI, the processing may be performed by the AI ​​in part or in whole, but is not limited to this example. Also, processing performed by an AI including a generative AI may be replaced by rule-based processing, and rule-based processing may be replaced by processing performed by an AI including a generative AI.

[0195] The data processing system 310 according to the third embodiment performs the same processing as the data processing system 10 according to the first embodiment. The processing by the data processing system 310 is performed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the headset terminal 314, but may also be performed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the headset terminal 314. In addition, the specific processing unit 290 of the data processing device 12 acquires or collects information necessary for processing from the headset terminal 314 or an external device, and the headset terminal 314 acquires or collects information necessary for processing from the data processing device 12 or an external device.

[0196] Each of the multiple elements described above, including the application acceptance unit, testing unit, review unit, certification unit, audit unit, robot instruction unit, robot evaluation unit, dialogue evaluation unit, and instruction analysis unit, is implemented by, for example, at least one of the headset terminal 314 and the data processing unit 12. For example, the application acceptance unit is implemented by the control unit 46A of the headset terminal 314 and can accept online applications. The testing unit is implemented by the specific processing unit 290 of the data processing unit 12 and tests the skills and ethical aspects of the AI ​​agent. The review unit is implemented by the specific processing unit 290 of the data processing unit 12 and reviews the test results. The certification unit is implemented by the specific processing unit 290 of the data processing unit 12 and performs certification. The audit unit is implemented by the specific processing unit 290 of the data processing unit 12 and audits the activities of the certified AI agent. The robot instruction unit is implemented by the control unit 46A of the headset terminal 314 and allows the AI ​​agent to issue instructions to the physical robot. The robot evaluation unit is implemented by the specific processing unit 290 of the data processing device 12 and evaluates the robot's movements. The dialogue evaluation unit is implemented by the specific processing unit 290 of the data processing device 12 and evaluates the dialogue capabilities of the AI ​​agent using generated AI. The instruction analysis unit is implemented by the specific processing unit 290 of the data processing device 12 and analyzes the instructions of the AI ​​agent using generated AI and issues appropriate operation instructions to the physical robot. The correspondence between each unit and the device or control unit is not limited to the example described above and can be modified in various ways.

[0197] [Fourth Embodiment] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[0198] As shown in Figure 7, the 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.

[0199] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN and / or LAN.

[0200] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[0201] The microphone 238 receives voice commands and other instructions from the user by receiving voice signals. The microphone 238 captures the voice signals from the user, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0202] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS image sensor or CCD image sensor, which captures images of the area around the user (for example, an imaging range defined by a field of view equivalent to the field of vision of a typical healthy person).

[0203] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0204] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive 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 robot 414's emotions can be expressed by controlling these motors. The robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[0205] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0206] The processor 28 reads a 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 acting as a specific processing unit 290 according to the specific processing program 56 executed on the RAM 30.

[0207] Storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290. The identification processing unit 290 can estimate the user's emotions using the emotion identification model 59 and perform identification processing using the user's emotions. The emotion estimation function (emotion identification function) using the emotion identification model 59 performs various estimations and predictions regarding the user's emotions, including but not limited to these examples. Furthermore, emotion estimation and prediction also include, for example, emotion analysis.

[0208] In robot 414, specific processing is performed by processor 46. A specific program 60 is stored in storage 50. Processor 46 reads the specific program 60 from storage 50 and executes it on RAM 48. The specific processing is achieved by processor 46 acting as a control unit 46A according to the specific program 60 executed on RAM 48. Robot 414 also has data generation model 58 and emotion identification model 59, similar to those of the robot, and can perform processing similar to that of the specific processing unit 290 using these models.

[0209] Furthermore, other devices besides the data processing device 12 may also have the data generation model 58. For example, a server device may have the data generation model 58. In this case, the data processing device 12 obtains processing results (such as prediction results) using the data generation model 58 by communicating with the server device that has the data generation model 58. Also, the data processing device 12 may be a server device or a terminal device owned by the user (for example, a mobile phone, robot, home appliance, etc.).

[0210] 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 controlled object 443 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0211] The data generation model 58 is a so-called generative AI. An example of a data generation model 58 is a generative AI such as ChatGPT. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and inference data such as audio data representing speech, text data representing text, and image data representing images (e.g., still image data or video data). The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference result in one or more data formats such as audio data, text data, and image data. The data generation model 58 includes, for example, text generation AI, image generation AI, and multimodal generation AI. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The specific processing unit 290 performs the specific processing described above using the data generation model 58. The data generation model 58 may be a fine-tuned model that outputs inference results from prompts that do not contain instructions, in which case the data generation model 58 can output inference results from prompts that do not contain instructions. In the data processing device 12, etc., there are multiple types of data generation models 58, and the data generation model 58 includes AI other than generative AI. AI other than generative AI includes, for example, linear regression, logistic regression, decision trees, random forests, support vector machines (SVM), k-means clustering, convolutional neural networks (CNN), recurrent neural networks (RNN), generative adversarial networks (GAN), or naive Bayes, and can perform various processes, but is not limited to these examples. Also, the AI ​​may be an AI agent. Furthermore, when the processing of each part described above is performed by the AI, the processing may be performed by the AI ​​in part or in whole, but is not limited to this example. Also, processing performed by an AI including a generative AI may be replaced by rule-based processing, and rule-based processing may be replaced by processing performed by an AI including a generative AI.

[0212] The data processing system 410 according to the fourth embodiment performs the same processing as the data processing system 10 according to the first embodiment. The processing by the data processing system 410 is performed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the robot 414, but it may also be performed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the robot 414. In addition, the specific processing unit 290 of the data processing device 12 acquires or collects information necessary for processing from the robot 414 or an external device, and the robot 414 acquires or collects information necessary for processing from the data processing device 12 or an external device.

[0213] Each of the multiple elements described above, including the application acceptance unit, testing unit, review unit, certification unit, audit unit, robot instruction unit, robot evaluation unit, dialogue evaluation unit, and instruction analysis unit, is implemented by, for example, at least one of the robot 414 and the data processing unit 12. For example, the application acceptance unit is implemented by the control unit 46A of the robot 414 and can accept online applications. The testing unit is implemented by the specific processing unit 290 of the data processing unit 12 and tests the skills and ethical aspects of the AI ​​agent. The review unit is implemented by the specific processing unit 290 of the data processing unit 12 and reviews the test results. The certification unit is implemented by the specific processing unit 290 of the data processing unit 12 and performs certification. The audit unit is implemented by the specific processing unit 290 of the data processing unit 12 and audits the activities of the certified AI agent. The robot instruction unit is implemented by the control unit 46A of the robot 414 and the AI ​​agent issues instructions to the physical robot. The robot evaluation unit is implemented by the specific processing unit 290 of the data processing unit 12 and evaluates the robot's operation. The dialogue evaluation unit is implemented by the specific processing unit 290 of the data processing device 12 and evaluates the dialogue capabilities of the AI ​​agent using generated AI. The instruction analysis unit is implemented by the specific processing unit 290 of the data processing device 12 and analyzes the instructions of the AI ​​agent using generated AI and issues appropriate operation instructions to the physical robot. The correspondence between each unit and the device and control unit is not limited to the example described above and can be modified in various ways.

[0214] Furthermore, the emotion identification model 59, acting 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 a specific mapping, which is an emotion map (see Figure 9). Similarly, the emotion identification model 59 may also determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[0215] Figure 9 shows the emotion map 400, in which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive the emotions are located. Further out of the concentric circles, emotions representing states and actions arising from mental states are located. Emotion is a concept that includes feelings and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions occurring in the brain are located. On the right side of the concentric circles, emotions that are generally induced by situational judgment are located. Above and below the concentric circles, emotions that are generally generated from reactions occurring in the brain and induced by situational judgment are located. In addition, the emotion of "pleasure" is located on the upper side of the concentric circles, and the emotion of "displeasure" is located on the lower side. Thus, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions arise, and emotions that are likely to occur simultaneously are mapped close together.

[0216] These emotions are distributed at the 3 o'clock position on the Emotion Map 400, and usually fluctuate between feelings of security and anxiety. In the right half of the Emotion Map 400, situational awareness takes precedence over internal feelings, resulting in a calm impression.

[0217] The inside of the Emotion Map 400 represents inner thoughts, while the outside represents actions. Therefore, the further you go from the outside of the Emotion Map 400, the more visible (expressed in actions) your emotions become.

[0218] Here, human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. Similarly, in robots, cars, and motorcycles, emotions can be created based on various balances, such as posture and battery level. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. The emotion map can be generated based, for example, on Dr. Mitsuyoshi's emotion map (Research on a system for analyzing brain physiological signals of speech emotion recognition and emotion, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map contains emotions belonging to a region called "response," where sensation is dominant. The right half of the emotion map contains emotions belonging to a region called "situation," where situational awareness is dominant.

[0219] The emotion map defines two emotions that promote learning. One is the emotion around the middle of the negative "repentance" and "reflection" on the situation side. In other words, it is 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 the emotion around the positive "desire" on the reaction side. In other words, it is when the robot has positive feelings such as "I want more" or "I want to know more."

[0220] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​representing each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple training data sets, which are combinations of user input and emotion values ​​representing each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions located close together have similar values, as shown in the emotion map 900 in Figure 10. Figure 10 shows an example where multiple emotions such as "reassured," "calm," and "confident" have similar emotion values.

[0221] In the above embodiment, an example was given in which a specific process is performed by a single computer 22. However, the technology of this disclosure is not limited thereto, and a distributed processing method for the specific process may be used, which includes computer 22 and multiple other computers.

[0222] In the above embodiment, an example was given in which the specific processing program 56 is stored in the storage 32, but the technology of this disclosure is not limited thereto. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-temporary storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-temporary storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes specific processing according to the specific processing program 56.

[0223] 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.

[0224] Furthermore, it is not necessary to store the entirety 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 the entirety of the specific processing program 56 in the storage 32; it is acceptable to store only a portion of the specific processing program 56.

[0225] The following types of processors can be used as hardware resources to perform specific processing. Examples of processors include a CPU, a general-purpose processor that functions as a hardware resource to perform specific processing by executing software, i.e., a program. Other examples of processors include dedicated electrical circuits, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices), or ASICs (Application Specific Integrated Circuits), which have circuit configurations specifically designed to perform specific processing. All of these processors have built-in or connected memory, and all of them perform specific processing by using memory.

[0226] The hardware resource that performs a specific process may consist of one of these various processors, or it may consist of 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). Alternatively, the hardware resource that performs a specific process may consist of a single processor.

[0227] Examples of configurations using a single processor include, firstly, a configuration in which one or more CPUs and software are combined to form a single processor, and this processor functions as a hardware resource that performs a specific process. Secondly, there is a configuration using a processor that realizes the functions of the entire system, including multiple hardware resources that perform a specific process, on a single IC chip, as exemplified by SoCs (System-on-a-chip). In this way, a specific process is realized using one or more of the above types of processors as hardware resources.

[0228] Furthermore, the hardware structure of these various processors can more specifically utilize electrical circuits that combine circuit elements such as semiconductor devices. Also, the specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps can be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose.

[0229] Furthermore, although the above-described examples were divided into four embodiments, some or all of these embodiments may be combined. Also, the smart device 14, smart glasses 214, headset terminal 314, and robot 414 are just examples, and they may be combined, or other devices may be used. Also, although the above-described examples were divided into two embodiments, Embodiment 1 and Embodiment 2, these may be combined.

[0230] The descriptions and illustrations presented above are detailed explanations of the technical aspects of this disclosure and are merely examples of the technical aspects. For example, the above descriptions of the structure, function, operation, and effect are examples of the structure, function, operation, and effect of the technical aspects of this disclosure. Therefore, it goes without saying that you may delete unnecessary parts, add new elements, or replace elements in the descriptions and illustrations presented above, as long as you do not deviate from the essence of the technical aspects of this disclosure. Furthermore, in order to avoid confusion and facilitate understanding of the technical aspects of this disclosure, explanations of common technical knowledge and other things that do not require special explanation to enable the implementation of the technical aspects of this disclosure have been omitted from the descriptions and illustrations presented above.

[0231] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted to be incorporated by reference.

[0232] (Note 1) The application acceptance department that receives applications, A testing unit that tests applications received by the aforementioned application acceptance unit, A review department that reviews the results of tests conducted by the aforementioned testing department, An accreditation department that grants accreditation based on the results of the review conducted by the aforementioned review department, The system includes an audit department that audits the activities of AI agents certified by the aforementioned certification department. A system characterized by the following features. (Note 2) To verify the behavioral and conversational capabilities of the AI ​​agent, it is equipped with a robot command unit that integrates with a physical robot. The system described in Appendix 1, characterized by the features described herein. (Note 3) The system includes a robot evaluation unit where an autonomous AI agent evaluates the robot's movements and verifies their accuracy and practicality. The system described in Appendix 2, characterized by the features described herein. (Note 4) It utilizes generative AI to conduct dialogues and includes a dialogue evaluation unit that evaluates the dialogue capabilities of the AI ​​agent during those dialogues. The system described in Appendix 2, characterized by the features described herein. (Note 5) The generated AI includes an instruction analysis unit that analyzes the instructions of the AI ​​agent and issues appropriate motion instructions to the physical robot. The system described in Appendix 2, characterized by the features described herein. (Note 6) The aforementioned application receiving department, The system estimates the user's emotions and adjusts the timing of application acceptance based on those estimated emotions. The system described in Appendix 1, characterized by the features described herein. (Note 7) The aforementioned application receiving department, We analyze past application data and select the optimal application acceptance method. The system described in Appendix 1, characterized by the features described herein. (Note 8) The aforementioned application receiving department, When an application is received, it will be filtered based on the applicant's current situation and areas of interest. The system described in Appendix 1, characterized by the features described herein. (Note 9) The aforementioned application receiving department, It estimates the user's emotions and determines the priority of applications to be accepted based on the estimated user emotions. The system described in Appendix 1, characterized by the features described herein. (Note 10) The aforementioned application receiving department, When receiving applications, priority will be given to accepting applications that are highly relevant, taking into account the applicant's geographical location. The system described in Appendix 1, characterized by the features described herein. (Note 11) The aforementioned application receiving department, Upon receiving an application, the applicant's social media activity will be analyzed, and relevant applications will be accepted. The system described in Appendix 1, characterized by the features described herein. (Note 12) The aforementioned test section is We estimate the user's emotions and adjust the presentation of the test based on those estimated emotions. The system described in Appendix 1, characterized by the features described herein. (Note 13) The aforementioned test section is During the test, the level of detail in the test will be adjusted based on the importance of the AI ​​agent. The system described in Appendix 1, characterized by the features described herein. (Note 14) The aforementioned test section is During testing, different test algorithms are applied depending on the category of the AI ​​agent. The system described in Appendix 1, characterized by the features described herein. (Note 15) The aforementioned test section is The system estimates the user's emotions and adjusts the length of the test based on those emotions. The system described in Appendix 1, characterized by the features described herein. (Note 16) The aforementioned test section is During the exam, the priority of the exam will be determined based on when the AI ​​agent submits their answers. The system described in Appendix 1, characterized by the features described herein. (Note 17) The aforementioned test section is During the test, the order of the tests will be adjusted based on the relevance of the AI ​​agents. The system described in Appendix 1, characterized by the features described herein. (Note 18) The aforementioned review department, We estimate the user's emotions and adjust the review criteria based on those estimated emotions. The system described in Appendix 1, characterized by the features described herein. (Note 19) The aforementioned review department, During the review process, we improve the accuracy of the review by considering the interrelationships of AI agents. The system described in Appendix 1, characterized by the features described herein. (Note 20) The aforementioned review department, During the review process, the AI ​​agent will take into consideration the applicant's attribute information. The system described in Appendix 1, characterized by the features described herein. (Note 21) The aforementioned review department, The system estimates the user's emotions and adjusts the order in which the review results are displayed based on those estimated emotions. The system described in Appendix 1, characterized by the features described herein. (Note 22) The aforementioned review department, During the review process, the geographical distribution of AI agents will be taken into consideration. The system described in Appendix 1, characterized by the features described herein. (Note 23) The aforementioned review department, During the review process, the AI ​​agent references relevant literature to improve the accuracy of the review. The system described in Appendix 1, characterized by the features described herein. (Note 24) The aforementioned certification department, We estimate the user's emotions and adjust the certification method based on the estimated user emotions. The system described in Appendix 1, characterized by the features described herein. (Note 25) The aforementioned certification department, During the certification process, the AI ​​agent's past activities are analyzed to select the most suitable certification method. The system described in Appendix 1, characterized by the features described herein. (Note 26) The aforementioned certification department, During certification, the certification process is customized based on the AI ​​agent's current status. The system described in Appendix 1, characterized by the features described herein. (Note 27) The aforementioned certification department, The system estimates the user's emotions and determines the priority of certifications based on those estimated emotions. The system described in Appendix 1, characterized by the features described herein. (Note 28) The aforementioned certification department, During the certification process, the optimal certification method will be selected, taking into account the geographical location information of the AI ​​agent. The system described in Appendix 1, characterized by the features described herein. (Note 29) The aforementioned certification department, During the certification process, we analyze the AI ​​agent's social media activity and propose certification methods. The system described in Appendix 1, characterized by the features described herein. (Note 30) The aforementioned audit department, We estimate user sentiment and adjust the audit method based on the estimated user sentiment. The system described in Appendix 1, characterized by the features described herein. (Note 31) The aforementioned audit department, During the audit, the AI ​​agent's past activities are analyzed to select the optimal audit method. The system described in Appendix 1, characterized by the features described herein. (Note 32) The aforementioned audit department, During the audit, customize the audit method based on the current status of the AI ​​agent. The system described in Appendix 1, characterized by the features described herein. (Note 33) The aforementioned audit department, Estimate the user's emotions and determine the priority of the audit based on the estimated user emotions The system according to Appendix 1, characterized by the above (Appendix 34) The audit department When conducting an audit, select the optimal audit method considering the geographical location information of the AI agent The system according to Appendix 1, characterized by the above (Appendix 35) The audit department When conducting an audit, analyze the social media activities of the AI agent and propose means of the audit The system according to Appendix 1, characterized by the above (Appendix 36) The robot instruction department Estimate the user's emotions and adjust the method of instructing the robot based on the estimated user emotions The system according to Appendix 2, characterized by the above (Appendix 37) The robot instruction department When instructing the robot, select the optimal instruction method by analyzing the past instruction history of the AI agent The system according to Appendix 2, characterized by the above (Appendix 38) The robot instruction department When instructing the robot, customize the means of instruction based on the current situation of the AI agent The system according to Appendix 2, characterized by the above (Appendix 39) The robot instruction department Estimate the user's emotions and determine the priority of the instruction to the robot based on the estimated user emotions The system according to Appendix 2, characterized by the above (Appendix 40) The robot instruction department When instructing the robot, select the optimal instruction method considering the geographical location information of the AI agent The system according to Appendix 2, characterized by the above (Note 41) The robot instruction unit is When giving instructions to a robot, the AI ​​agent analyzes its social media activity and suggests ways to give instructions. The system described in Appendix 2, characterized by the features described herein. (Note 42) The robot evaluation unit is It estimates the user's emotions and adjusts the robot's evaluation method based on the estimated user emotions. The system described in Appendix 3, characterized by the features described herein. (Note 43) The robot evaluation unit is During robot evaluation, the AI ​​agent's past evaluation history is analyzed to select the optimal evaluation method. The system described in Appendix 3, characterized by the features described herein. (Note 44) The robot evaluation unit is During robot evaluation, the evaluation method is customized based on the current status of the AI ​​agent. The system described in Appendix 3, characterized by the features described herein. (Note 45) The robot evaluation unit is It estimates the user's emotions and determines the priority of the robot's evaluation based on the estimated user emotions. The system described in Appendix 3, characterized by the features described herein. (Note 46) The robot evaluation unit is When evaluating robots, the optimal evaluation method is selected by considering the geographical location information of the AI ​​agent. The system described in Appendix 3, characterized by the features described herein. (Note 47) The robot evaluation unit is When evaluating robots, we propose evaluation methods by analyzing the social media activity of AI agents. The system described in Appendix 3, characterized by the features described herein. (Note 48) The aforementioned dialogue evaluation unit, It estimates the user's emotions and adjusts the interaction evaluation method based on the estimated user emotions. The system according to appended note 4, characterized in that... (Appended note 49) The dialogue evaluation unit selects an optimal evaluation method by analyzing the past dialogue history of the AI agent during dialogue evaluation The system according to appended note 4, characterized in that... (Appended note 50) The dialogue evaluation unit customizes the evaluation means based on the current situation of the AI agent during dialogue evaluation The system according to appended note 4, characterized in that... (Appended note 51) The dialogue evaluation unit estimates the user's emotion and determines the priority order of dialogue evaluation based on the estimated user's emotion The system according to appended note 4, characterized in that... (Appended note 52) The dialogue evaluation unit selects an optimal evaluation method by considering the geographical location information of the AI agent during dialogue evaluation The system according to appended note 4, characterized in that... (Appended note 53) The dialogue evaluation unit proposes evaluation means by analyzing the social media activities of the AI agent during dialogue evaluation The system according to appended note 4, characterized in that... (Appended note 54) The instruction analysis unit estimates the user's emotion and adjusts the instruction analysis method based on the estimated user's emotion The system according to appended note 5, characterized in that... (Appended note 55) The instruction analysis unit selects an optimal analysis method by analyzing the past instruction history of the AI agent during instruction analysis The system according to appended note 5, characterized in that... (Appended note 56) The instruction analysis unit During instruction analysis, the analysis method is customized based on the current status of the AI ​​agent. The system described in Appendix 5, characterized by the features described herein. (Note 57) The instruction analysis unit, The system estimates the user's emotions and determines the priority of instruction analysis based on the estimated user emotions. The system described in Appendix 5, characterized by the features described herein. (Note 58) The instruction analysis unit, During instruction analysis, the optimal analysis method is selected by considering the geographical location information of the AI ​​agent. The system described in Appendix 5, characterized by the features described herein. (Note 59) The instruction analysis unit, During instruction analysis, the AI ​​agent's social media activity is analyzed to propose methods for analysis. The system described in Appendix 5, characterized by the features described herein. [Explanation of Symbols]

[0233] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots

Claims

1. The application acceptance department that receives applications, A testing unit that tests applications received by the aforementioned application acceptance unit, A review department that reviews the results of tests conducted by the aforementioned testing department, An accreditation department that grants accreditation based on the results of the review conducted by the aforementioned review department, The system includes an audit department that audits the activities of AI agents certified by the aforementioned certification department. A system characterized by the following features.

2. To verify the AI ​​agent's behavioral and conversational capabilities, it is equipped with a robot command unit that integrates with a physical robot. The system according to feature 1.

3. The system includes a robot evaluation unit where an autonomous AI agent evaluates the robot's movements and verifies their accuracy and practicality. The system according to feature 2.

4. It includes a dialogue evaluation unit that uses generative AI to conduct conversations and evaluates the dialogue capabilities of the AI ​​agent during those conversations. The system according to feature 2.

5. The generating AI includes an instruction analysis unit that analyzes the instructions of the AI ​​agent and issues appropriate motion instructions to the physical robot. The system according to feature 2.

6. The aforementioned application receiving department, The system estimates the user's emotions and adjusts the timing of application acceptance based on those estimated emotions. The system according to feature 1.

7. The aforementioned application receiving department, We analyze past application data and select the optimal application acceptance method. The system according to feature 1.

8. The aforementioned application receiving department, When an application is received, it will be filtered based on the applicant's current situation and areas of interest. The system according to feature 1.

9. The aforementioned application receiving department, It estimates the user's emotions and determines the priority of applications to be accepted based on the estimated user emotions. The system according to feature 1.

Images