system

Abstract

The system according to this embodiment aims for the AI ​​to comply with ethical guidelines and constraints and to exercise self-control. [Solution] The system according to the embodiment comprises a listing unit, a rule setting unit, a reward / penalty unit, a feedback unit, and an ethical judgment unit. The listing unit lists permitted and prohibited actions. The rule setting unit sets actions that are permitted only under specific circumstances. The reward / penalty unit rewards desirable actions and penalizes undesirable actions. The feedback unit adjusts the algorithm based on the AI's actions. The ethical judgment unit makes ethical judgments.

Description

Technical Field

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

Background Art

[0002] Patent Document 1 discloses a method for controlling a persona chatbot, which is performed by at least one processor and includes steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to an explanation of a character of the chatbot, 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 prior art, AI has no ethical guidelines or constraints and may take unexpected actions.

[0005] The system according to the embodiment aims to make AI comply with ethical guidelines and constraints and perform self-control.

Means for Solving the Problems

[0006] The system according to this embodiment comprises a listing unit, a rule setting unit, a reward / penalty unit, a feedback unit, and an ethical judgment unit. The listing unit lists permitted and prohibited actions. The rule setting unit sets actions that are permitted only under specific circumstances. The reward / penalty unit rewards desirable actions and penalizes undesirable actions. The feedback unit adjusts the algorithm based on the AI's actions. The ethical judgment unit makes ethical judgments. [Effects of the Invention]

[0007] The system according to this embodiment allows the AI ​​to comply with ethical guidelines and constraints and to exercise self-control. [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 labeled 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 control system according to an embodiment of the present invention is a system that provides a mechanism for a generative AI to comply with ethical guidelines and exercise self-control. This AI control system enables the generative AI to comply with ethical guidelines, exercise self-control, and suppress undesirable behavior on its own. For example, the AI ​​control system sets up a whitelist / blacklist for the AI, listing permitted and prohibited behaviors. Next, it introduces conditional rules that set behaviors permitted only under specific circumstances. It also constructs a mechanism that rewards desirable behavior and penalizes undesirable behavior using reinforcement learning. Furthermore, it introduces a feedback system that adjusts the algorithm based on the AI's behavior results and adds a module for making ethical judgments. It ensures transparency and accountability to clarify the AI's behavioral decisions and provides a monitoring system in which humans monitor the AI's behavior and intervene when problems occur. It periodically reviews the AI's behavior and decisions and makes adjustments as necessary. It also introduces an emergency stop function that immediately stops the system when undesirable behavior occurs, and a fail-safe system that ensures safety even in the event of malfunction. As a result, the AI ​​control system enables the generative AI to comply with ethical guidelines, exercise self-control, and suppress undesirable behavior on its own. This will improve the reliability of AI and earn the trust of society as a whole. As a result, AI control systems will be able to ensure that generated AI adheres to ethical guidelines, exercises self-regulation, and suppresses undesirable behavior on its own.

[0029] The AI ​​control system according to this embodiment comprises a listing unit, a rule setting unit, a reward / penalty unit, a feedback unit, and an ethical judgment unit. The listing unit lists permitted and prohibited actions. The listing unit, for example, sets a whitelist / blacklist for the AI ​​and lists permitted and prohibited actions. The listing unit predefines the actions the AI ​​will perform and controls it to perform only permitted actions. The rule setting unit sets actions that are permitted only under specific circumstances. The rule setting unit restricts actions based on, for example, specific time periods, locations, environmental conditions, etc. The rule setting unit controls the AI ​​to perform actions only when specific conditions are met. The reward / penalty unit rewards desirable actions and penalizes undesirable actions. The reward / penalty unit evaluates the AI's actions using, for example, reinforcement learning, rewards desirable actions, and penalizes undesirable actions. The reward / penalty unit controls the AI ​​to reinforce desirable actions and suppress undesirable actions through learning. The feedback unit adjusts the algorithm based on the AI's actions. The feedback unit monitors the AI's actions and adjusts the algorithm parameters, for example. The feedback unit optimizes the algorithm so that the AI ​​takes appropriate actions. The ethical judgment unit makes ethical judgments. The ethical judgment unit evaluates whether the actions taken by the AI ​​comply with ethical guidelines, for example. The ethical judgment unit controls the AI ​​to take ethical actions. As a result, the AI ​​control system according to the embodiment can enable the generated AI to comply with ethical guidelines, exercise self-control, and suppress undesirable actions on its own.

[0030] The listing unit lists permitted and prohibited actions. For example, the listing unit sets whitelists and blacklists for AI, listing permitted and prohibited actions. Specifically, the whitelist includes actions and operations that the AI ​​can perform, and the blacklist includes actions and operations that are prohibited. This allows for pre-defining the actions that the AI ​​will perform and controlling it to perform only permitted actions. The listing unit categorizes the AI's actions in detail and sets a permitted or prohibited flag for each action. For example, if the AI ​​collects data from the internet, it can be allowed to access only specific websites and prohibited from accessing other websites. The listing unit also has the function to dynamically update the list, and can change the whitelist and blacklist as needed. For example, if a new security risk is discovered, it can be immediately added to the blacklist, restricting the AI's actions. Furthermore, the listing unit provides an interface for users and administrators to manually edit the list, enabling flexible management. This allows the listing unit to strictly control the AI's actions and prevent unexpected or inappropriate behavior.

[0031] The rule-setting unit sets actions that are permitted only under specific circumstances. For example, it restricts actions based on specific time periods, locations, environmental conditions, etc. Specifically, it controls the AI ​​to perform actions only when certain conditions are met. For example, it is possible to prohibit the AI ​​from performing certain operations at night and only allow them during the day. It is also possible to permit actions only within a specific geographical range and restrict actions outside of that range. The rule-setting unit utilizes environmental sensors and GPS data to enable the AI ​​to accurately understand the current situation and control its actions according to the set rules. Furthermore, the rule-setting unit can also set complex rules that combine multiple conditions. For example, it is possible to set a rule that permits a specific action only when certain temperature and humidity conditions are met. This allows the rule-setting unit to flexibly and precisely control the AI's actions and minimize risks under specific circumstances.

[0032] The reward-penalty unit rewards desirable behaviors and penalizes undesirable behaviors. For example, it uses reinforcement learning to evaluate the AI's behavior, rewarding desirable actions and penalizing undesirable ones. Specifically, it rewards the AI ​​with points or virtual currency when it takes appropriate actions, and penalizes it with points reductions or behavioral restrictions when it takes inappropriate actions. The reward-penalty unit controls the AI ​​so that it reinforces desirable behaviors and suppresses undesirable behaviors through learning. For example, by rewarding the AI ​​when it takes actions that protect user privacy and penalizing it when it takes actions that infringe on privacy, the AI ​​learns to prioritize privacy protection. The reward-penalty unit also has the function of dynamically adjusting the criteria for rewards and penalties, and can set optimal evaluation criteria according to the AI's learning progress and changes in the environment. In this way, the reward-penalty unit can effectively control the AI's behavior and promote desirable actions.

[0033] The feedback unit adjusts the algorithm based on the AI's actions. For example, the feedback unit monitors the AI's actions and adjusts the algorithm's parameters. Specifically, it evaluates the results of the AI's actions and adjusts the algorithm's weights and biases based on that evaluation. For example, if the AI ​​successfully completes a particular task, it analyzes the factors contributing to that success and reflects them in the algorithm. If it fails, it identifies the cause and modifies the algorithm to prevent the same failure from happening again. The feedback unit can monitor the AI's actions in real time and quickly optimize the algorithm. Furthermore, the feedback unit can also collect feedback from users and reflect their evaluation of the AI's actions. For example, it can conduct a survey where users rate their satisfaction with the AI's actions and adjust the algorithm based on the results. This allows the feedback unit to optimize the algorithm so that the AI ​​takes appropriate actions and improve the overall system performance.

[0034] The Ethics Department makes ethical judgments. For example, it evaluates whether the actions of an AI comply with ethical guidelines. Specifically, it evaluates whether the actions of an AI respect human rights, protect privacy, and are socially acceptable. The Ethics Department sets ethical guidelines in advance to control the AI ​​to act ethically and evaluates the AI's actions based on those guidelines. For example, when an AI handles personal information, it verifies that it is handling it appropriately from a privacy protection standpoint. Also, when an AI makes decisions, it evaluates whether fairness and transparency are ensured and whether there is any bias towards specific individuals or groups. Furthermore, the Ethics Department evaluates whether the AI's actions are socially acceptable and makes judgments that take social impact into consideration. In this way, the Ethics Department can control the AI ​​to comply with ethical guidelines and act socially responsible.

[0035] The AI ​​control system includes a monitoring system. The monitoring system allows humans to monitor the AI's behavior and intervene when problems occur. The monitoring system monitors the AI's behavior in real time, for example, using cameras and sensors. The monitoring system records the AI's behavior and issues alerts if an anomaly is detected. The monitoring system saves the AI's behavior logs for later review. The monitoring system includes dedicated monitoring software for monitoring the AI's behavior. This allows humans to monitor the AI's behavior and intervene when problems occur. Some or all of the above processes in the monitoring system may be performed using AI, for example, or without AI. For example, the monitoring system can input video data acquired by cameras into a generating AI and have the generating AI perform anomaly detection.

[0036] The AI ​​control system is equipped with an emergency stop function. The emergency stop function can immediately stop the system if undesirable behavior occurs. The emergency stop function stops the AI's operation, for example, by button operation or sensor detection. The emergency stop function activates automatically when the AI ​​takes abnormal behavior. The emergency stop function is equipped with software control to safely stop the AI's operation. This allows the emergency stop function to immediately stop the system if undesirable behavior occurs. Some or all of the above-described processes in the emergency stop function may be performed using AI, for example, or without AI. For example, the emergency stop function can input abnormal data detected by sensors into a generating AI and have the generating AI make a decision to perform an emergency stop.

[0037] The AI ​​control system includes a fail-safe system. The fail-safe system ensures safety even in the event of a malfunction. The fail-safe system includes, for example, redundant systems and backup functions. The fail-safe system automatically transitions to a safe state if the AI ​​malfunctions. The fail-safe system is equipped with safety devices that activate when an AI malfunction is detected. This ensures safety even in the event of a malfunction. Some or all of the above-described processes in the fail-safe system may be performed using, for example, the AI, or not using the AI. For example, the fail-safe system can have the generating AI perform the switching of the redundant system.

[0038] The listing unit can analyze past behavioral history and improve the accuracy of the behaviors listed. For example, the listing unit can prioritize displaying behaviors that the user has frequently performed in the past when listing them. The listing unit can also predict and list behaviors to be performed during specific time periods based on the user's past behavioral history. The listing unit can also analyze the user's past behavioral patterns and list the most efficient behaviors. This allows for improved accuracy of behavior based on past behavioral history. Some or all of the above-described processes in the listing unit may be performed using AI, for example, or without AI. For example, the listing unit can input past behavioral data into a generating AI and have the generating AI perform behavioral predictions.

[0039] The listing unit can dynamically update the actions to be listed according to specific time periods or situations. For example, the listing unit can list actions that a user performs in the morning and different actions in the afternoon and evening. The listing unit can also list actions appropriate for a specific location if the user is in that location. The listing unit can also dynamically update the actions to be listed according to the user's current situation (e.g., at work or on vacation). This allows for dynamic updates of actions according to time periods and situations. Some or all of the above processing in the listing unit may be performed using AI, for example, or without AI. For example, the listing unit can input the user's current situation data into a generating AI and have the generating AI perform the action update.

[0040] The rule setting unit can analyze past rule application history and improve the accuracy of rules. For example, the rule setting unit can analyze the results of rules previously applied by the user and set the optimal rule. For example, the rule setting unit can predict and adjust rule application in specific situations based on the user's past rule application history. For example, the rule setting unit can analyze the user's past rule application patterns and set the most effective rule. This allows for improved rule accuracy based on past rule application history. Some or all of the above processes in the rule setting unit may be performed using AI, for example, or without AI. For example, the rule setting unit can input past rule application data into a generating AI and have the generating AI perform rule optimization.

[0041] The rule setting unit can dynamically change rules according to specific time periods or situations. For example, the rule setting unit can set rules to apply to the user during the morning hours and different rules to apply during the day and at night. The rule setting unit can also set rules appropriate for a specific location if the user is in that location. The rule setting unit can also dynamically change rules according to the user's current situation (e.g., at work or on vacation). This allows for dynamic rule changes according to time periods and situations. Some or all of the above-described processes in the rule setting unit may be performed using AI, for example, or without AI. For example, the rule setting unit can input the user's current situation data into a generating AI and have the generating AI execute rule changes.

[0042] The reward and penalty unit can analyze past reward and penalty history to improve the accuracy of rewards and penalties. For example, the reward and penalty unit can analyze the results of rewards and penalties a user has received in the past and set the optimal reward and penalty. For example, the reward and penalty unit can predict and adjust rewards and penalties in specific situations based on a user's past reward and penalty history. For example, the reward and penalty unit can analyze a user's past reward and penalty patterns and set the most effective reward and penalty. This allows for improved accuracy of rewards and penalties based on past reward and penalty history. Some or all of the above processing in the reward and penalty unit may be performed using AI, for example, or without AI. For example, the reward and penalty unit can input past reward and penalty data into a generating AI and have the generating AI perform the optimization of rewards and penalties.

[0043] The reward / penalty unit can dynamically change rewards and penalties according to specific time periods or situations. For example, the reward / penalty unit can set rewards and penalties for a user in the morning and apply different rewards and penalties during the day and night. The reward / penalty unit can also set rewards and penalties appropriate to a user's location if the user is in a specific place. The reward / penalty unit can also dynamically change rewards and penalties according to the user's current situation (e.g., at work or on vacation). This allows for dynamic changes to rewards and penalties according to time periods and situations. Some or all of the above-described processes in the reward / penalty unit may be performed using AI, for example, or without AI. For example, the reward / penalty unit can input the user's current situation data into a generating AI and have the generating AI execute the changes to rewards and penalties.

[0044] The feedback unit can analyze past feedback history and improve the accuracy of feedback. For example, the feedback unit can analyze the results of feedback the user has received in the past and set the optimal feedback. The feedback unit can also predict and adjust feedback in specific situations based on the user's past feedback history. For example, the feedback unit can analyze the user's past feedback patterns and set the most effective feedback. This allows for improved feedback accuracy based on past feedback history. Some or all of the above processes in the feedback unit may be performed using AI, for example, or without AI. For example, the feedback unit can input past feedback data into a generating AI and have the generating AI perform feedback optimization.

[0045] The feedback unit can dynamically change the feedback depending on the time of day or situation. For example, the feedback unit can set the feedback a user receives in the morning and apply different feedback during the day and at night. For example, the feedback unit can also set feedback appropriate for a specific location if the user is in a particular place. For example, the feedback unit can dynamically change the feedback depending on the user's current situation (e.g., working or on vacation). This allows for dynamic changes to the feedback depending on the time of day or situation. Some or all of the above processing in the feedback unit may be performed using AI, for example, or without AI. For example, the feedback unit can input the user's current situation data into a generating AI and have the generating AI perform the feedback changes.

[0046] The ethical judgment unit can analyze past ethical judgment history and improve the accuracy of ethical judgments. For example, the ethical judgment unit can analyze the results of ethical judgments made by the user in the past and set the optimal ethical judgment. For example, the ethical judgment unit can predict and adjust ethical judgments in specific situations based on the user's past ethical judgment history. For example, the ethical judgment unit can analyze the user's past ethical judgment patterns and set the most effective ethical judgment. This allows for improved accuracy of ethical judgments based on past ethical judgment history. Some or all of the above processes in the ethical judgment unit may be performed using AI, for example, or without AI. For example, the ethical judgment unit can input past ethical judgment data into a generating AI and have the generating AI perform the optimization of ethical judgments.

[0047] The ethical judgment unit can dynamically change ethical judgments according to specific time periods or circumstances. For example, the ethical judgment unit can set ethical judgments for the user in the morning and apply different ethical judgments during the day and at night. For example, the ethical judgment unit can also set ethical judgments appropriate to a specific location if the user is in that location. For example, the ethical judgment unit can dynamically change ethical judgments according to the user's current situation (e.g., at work or on vacation). This allows for dynamic changes to ethical judgments depending on the time period and circumstances. Some or all of the above processing in the ethical judgment unit may be performed using AI, for example, or without AI. For example, the ethical judgment unit can input the user's current situation data into a generating AI and have the generating AI perform the change in ethical judgments.

[0048] The monitoring system can analyze past monitoring history to improve monitoring accuracy. For example, the monitoring system can analyze the results of past monitoring received by the user and set optimal monitoring. For example, the monitoring system can predict and adjust monitoring in specific situations based on the user's past monitoring history. For example, the monitoring system can analyze the user's past monitoring patterns and set the most effective monitoring. This allows for improved monitoring accuracy based on past monitoring history. Some or all of the above processes in the monitoring system may be performed using AI, for example, or without AI. For example, the monitoring system can input past monitoring data into a generating AI and have the generating AI perform monitoring optimization.

[0049] The emergency stop function can analyze past emergency stop history to improve the accuracy of emergency stops. For example, the emergency stop function can analyze the results of emergency stops performed by the user in the past and set the optimal emergency stop. For example, the emergency stop function can predict and adjust emergency stops in specific situations based on the user's past emergency stop history. For example, the emergency stop function can analyze the user's past emergency stop patterns and set the most effective emergency stop. This allows for improved accuracy of emergency stops based on past emergency stop history. Some or all of the above processes in the emergency stop function may be performed using AI, for example, or without AI. For example, the emergency stop function can input past emergency stop data into a generating AI and have the generating AI perform emergency stop optimization.

[0050] A fail-safe system can analyze past fail-safe history to improve the accuracy of its fail-safes. For example, a fail-safe system can analyze the results of past fail-safes performed by a user and set the optimal fail-safe. A fail-safe system can also predict and adjust fail-safes in specific situations based on a user's past fail-safe history. A fail-safe system can also analyze a user's past fail-safe patterns and set the most effective fail-safe. This allows for improved fail-safe accuracy based on past fail-safe history. Some or all of the above processes in a fail-safe system may be performed using AI, for example, or without AI. For example, a fail-safe system can input past fail-safe data into a generating AI and have the generating AI perform fail-safe optimization.

[0051] The failsafe system can customize its failsafes based on the user's geographical location. For example, if the user is in a specific city, the failsafe system can set up failsafes to apply in that city. For example, if the user is traveling, the failsafe system can also set up failsafes to apply at the travel destination. For example, if the user is at home, the failsafe system can also set up failsafes to apply at home. This allows the failsafe to be customized based on the user's geographical location. Some or all of the above processes in the failsafe system may be performed using AI, for example, or not using AI. For example, the failsafe system can input the user's geographical location data into a generating AI and have the generating AI perform the failsafe customization.

[0052] The system according to the embodiment is not limited to the example described above, and various modifications are possible, for example, as follows.

[0053] The AI ​​control system can analyze past behavioral history and improve the accuracy of the actions listed. For example, it can prioritize displaying actions that a user has frequently performed in the past when listing them. It can also predict and list actions that a user will perform during specific time periods based on their past behavioral history. It can also analyze a user's past behavioral patterns and list the most efficient actions. This allows for improved accuracy of actions based on past behavioral history. Some or all of the above processing in the listing unit may be performed using AI or not. For example, the listing unit can input past behavioral data into a generating AI and have the generating AI perform behavioral predictions.

[0054] The AI ​​control system can dynamically update the list of actions according to specific time periods and situations. For example, it can list actions a user takes in the morning and different actions for the afternoon and evening. If the user is in a specific location, it can also list actions appropriate for that location. It can also dynamically update the list of actions according to the user's current situation (e.g., at work or on vacation). This allows for dynamic updates of actions according to time periods and situations. Some or all of the above processing in the listing unit may be performed using AI or not. For example, the listing unit can input the user's current situation data into a generating AI and have the generating AI perform the action update.

[0055] The AI ​​control system can analyze past rule application history and improve the accuracy of rules. For example, it can analyze the results of rules previously applied by the user and set the optimal rules. It can also predict and adjust rule application in specific situations based on the user's past rule application history. It can also analyze the user's past rule application patterns and set the most effective rules. This allows for improved rule accuracy based on past rule application history. Some or all of the above processing in the rule setting unit may be performed using AI or not. For example, the rule setting unit can input past rule application data into a generating AI and have the generating AI perform rule optimization.

[0056] The AI ​​control system can dynamically change rules according to specific time periods or situations. For example, it can set rules to apply to the user in the morning and different rules to apply during the day and at night. If the user is in a specific location, it can also set rules appropriate for that location. It can also dynamically change rules according to the user's current situation (e.g., working or on vacation). This allows for dynamic rule changes according to time periods and situations. Some or all of the above-described processes in the rule setting unit may be performed using AI or not. For example, the rule setting unit can input user's current situation data into a generating AI and have the generating AI execute rule changes.

[0057] The AI ​​control system can analyze past reward and penalty history to improve the accuracy of rewards and penalties. For example, it can analyze the results of rewards and penalties a user has received in the past and set the optimal rewards and penalties. It can also predict and adjust rewards and penalties in specific situations based on the user's past reward and penalty history. It can also analyze the user's past reward and penalty patterns and set the most effective rewards and penalties. This allows for improved accuracy of rewards and penalties based on past reward and penalty history. Some or all of the above processing in the reward and penalty unit may be performed using AI or not. For example, the reward and penalty unit can input past reward and penalty data into a generating AI and have the generating AI perform the optimization of rewards and penalties.

[0058] The following briefly describes the processing flow for example form 1.

[0059] Step 1: The listing unit lists permitted and prohibited actions. For example, it sets up a whitelist / blacklist for the AI ​​and lists permitted and prohibited actions. The listing unit pre-defines the actions the AI ​​will perform and controls it to execute only permitted actions. Step 2: The rule setting unit sets actions that are permitted only under specific circumstances. For example, it restricts actions based on specific time periods, locations, or environmental conditions. The rule setting unit controls the AI ​​to perform actions only when specific conditions are met. Step 3: The reward-penalty unit rewards desirable behaviors and penalizes undesirable behaviors. For example, reinforcement learning is used to evaluate the AI's behavior, rewarding desirable behaviors and penalizing undesirable behaviors. The reward-penalty unit controls the AI ​​so that it reinforces desirable behaviors and suppresses undesirable behaviors through learning. Step 4: The feedback unit adjusts the algorithm based on the AI's actions. For example, it monitors the AI's actions and adjusts the algorithm's parameters. The feedback unit optimizes the algorithm so that the AI ​​takes appropriate actions. Step 5: The Ethics Judgment Unit makes ethical judgments. For example, it evaluates whether the actions taken by the AI ​​comply with ethical guidelines. The Ethics Judgment Unit controls the AI ​​to act ethically.

[0060] (Example of form 2) An AI control system according to an embodiment of the present invention is a system that provides a mechanism for a generative AI to comply with ethical guidelines and exercise self-control. This AI control system enables the generative AI to comply with ethical guidelines, exercise self-control, and suppress undesirable behavior on its own. For example, the AI ​​control system sets up a whitelist / blacklist for the AI, listing permitted and prohibited behaviors. Next, it introduces conditional rules that set behaviors permitted only under specific circumstances. It also constructs a mechanism that rewards desirable behavior and penalizes undesirable behavior using reinforcement learning. Furthermore, it introduces a feedback system that adjusts the algorithm based on the AI's behavior results and adds a module for making ethical judgments. It ensures transparency and accountability to clarify the AI's behavioral decisions and provides a monitoring system in which humans monitor the AI's behavior and intervene when problems occur. It periodically reviews the AI's behavior and decisions and makes adjustments as necessary. It also introduces an emergency stop function that immediately stops the system when undesirable behavior occurs, and a fail-safe system that ensures safety even in the event of malfunction. As a result, the AI ​​control system enables the generative AI to comply with ethical guidelines, exercise self-control, and suppress undesirable behavior on its own. This will improve the reliability of AI and earn the trust of society as a whole. As a result, AI control systems will be able to ensure that generated AI adheres to ethical guidelines, exercises self-regulation, and suppresses undesirable behavior on its own.

[0061] The AI ​​control system according to this embodiment comprises a listing unit, a rule setting unit, a reward / penalty unit, a feedback unit, and an ethical judgment unit. The listing unit lists permitted and prohibited actions. The listing unit, for example, sets a whitelist / blacklist for the AI ​​and lists permitted and prohibited actions. The listing unit predefines the actions the AI ​​will perform and controls it to perform only permitted actions. The rule setting unit sets actions that are permitted only under specific circumstances. The rule setting unit restricts actions based on, for example, specific time periods, locations, environmental conditions, etc. The rule setting unit controls the AI ​​to perform actions only when specific conditions are met. The reward / penalty unit rewards desirable actions and penalizes undesirable actions. The reward / penalty unit evaluates the AI's actions using, for example, reinforcement learning, rewards desirable actions, and penalizes undesirable actions. The reward / penalty unit controls the AI ​​to reinforce desirable actions and suppress undesirable actions through learning. The feedback unit adjusts the algorithm based on the AI's actions. The feedback unit monitors the AI's actions and adjusts the algorithm parameters, for example. The feedback unit optimizes the algorithm so that the AI ​​takes appropriate actions. The ethical judgment unit makes ethical judgments. The ethical judgment unit evaluates whether the actions taken by the AI ​​comply with ethical guidelines, for example. The ethical judgment unit controls the AI ​​to take ethical actions. As a result, the AI ​​control system according to the embodiment can enable the generated AI to comply with ethical guidelines, exercise self-control, and suppress undesirable actions on its own.

[0062] The listing unit lists permitted and prohibited actions. For example, the listing unit sets whitelists and blacklists for AI, listing permitted and prohibited actions. Specifically, the whitelist includes actions and operations that the AI ​​can perform, and the blacklist includes actions and operations that are prohibited. This allows for pre-defining the actions that the AI ​​will perform and controlling it to perform only permitted actions. The listing unit categorizes the AI's actions in detail and sets a permitted or prohibited flag for each action. For example, if the AI ​​collects data from the internet, it can be allowed to access only specific websites and prohibited from accessing other websites. The listing unit also has the function to dynamically update the list, and can change the whitelist and blacklist as needed. For example, if a new security risk is discovered, it can be immediately added to the blacklist, restricting the AI's actions. Furthermore, the listing unit provides an interface for users and administrators to manually edit the list, enabling flexible management. This allows the listing unit to strictly control the AI's actions and prevent unexpected or inappropriate behavior.

[0063] The rule-setting unit sets actions that are permitted only under specific circumstances. For example, it restricts actions based on specific time periods, locations, environmental conditions, etc. Specifically, it controls the AI ​​to perform actions only when certain conditions are met. For example, it is possible to prohibit the AI ​​from performing certain operations at night and only allow them during the day. It is also possible to permit actions only within a specific geographical range and restrict actions outside of that range. The rule-setting unit utilizes environmental sensors and GPS data to enable the AI ​​to accurately understand the current situation and control its actions according to the set rules. Furthermore, the rule-setting unit can also set complex rules that combine multiple conditions. For example, it is possible to set a rule that permits a specific action only when certain temperature and humidity conditions are met. This allows the rule-setting unit to flexibly and precisely control the AI's actions and minimize risks under specific circumstances.

[0064] The reward-penalty unit rewards desirable behaviors and penalizes undesirable behaviors. For example, it uses reinforcement learning to evaluate the AI's behavior, rewarding desirable actions and penalizing undesirable ones. Specifically, it rewards the AI ​​with points or virtual currency when it takes appropriate actions, and penalizes it with points reductions or behavioral restrictions when it takes inappropriate actions. The reward-penalty unit controls the AI ​​so that it reinforces desirable behaviors and suppresses undesirable behaviors through learning. For example, by rewarding the AI ​​when it takes actions that protect user privacy and penalizing it when it takes actions that infringe on privacy, the AI ​​learns to prioritize privacy protection. The reward-penalty unit also has the function of dynamically adjusting the criteria for rewards and penalties, and can set optimal evaluation criteria according to the AI's learning progress and changes in the environment. In this way, the reward-penalty unit can effectively control the AI's behavior and promote desirable actions.

[0065] The feedback unit adjusts the algorithm based on the AI's actions. For example, the feedback unit monitors the AI's actions and adjusts the algorithm's parameters. Specifically, it evaluates the results of the AI's actions and adjusts the algorithm's weights and biases based on that evaluation. For example, if the AI ​​successfully completes a particular task, it analyzes the factors contributing to that success and reflects them in the algorithm. If it fails, it identifies the cause and modifies the algorithm to prevent the same failure from happening again. The feedback unit can monitor the AI's actions in real time and quickly optimize the algorithm. Furthermore, the feedback unit can also collect feedback from users and reflect their evaluation of the AI's actions. For example, it can conduct a survey where users rate their satisfaction with the AI's actions and adjust the algorithm based on the results. This allows the feedback unit to optimize the algorithm so that the AI ​​takes appropriate actions and improve the overall system performance.

[0066] The Ethics Department makes ethical judgments. For example, it evaluates whether the actions of an AI comply with ethical guidelines. Specifically, it evaluates whether the actions of an AI respect human rights, protect privacy, and are socially acceptable. The Ethics Department sets ethical guidelines in advance to control the AI ​​to act ethically and evaluates the AI's actions based on those guidelines. For example, when an AI handles personal information, it verifies that it is handling it appropriately from a privacy protection standpoint. Also, when an AI makes decisions, it evaluates whether fairness and transparency are ensured and whether there is any bias towards specific individuals or groups. Furthermore, the Ethics Department evaluates whether the AI's actions are socially acceptable and makes judgments that take social impact into consideration. In this way, the Ethics Department can control the AI ​​to comply with ethical guidelines and act socially responsible.

[0067] The AI ​​control system includes a monitoring system. The monitoring system allows humans to monitor the AI's behavior and intervene when problems occur. The monitoring system monitors the AI's behavior in real time, for example, using cameras and sensors. The monitoring system records the AI's behavior and issues alerts if an anomaly is detected. The monitoring system saves the AI's behavior logs for later review. The monitoring system includes dedicated monitoring software for monitoring the AI's behavior. This allows humans to monitor the AI's behavior and intervene when problems occur. Some or all of the above processes in the monitoring system may be performed using AI, for example, or without AI. For example, the monitoring system can input video data acquired by cameras into a generating AI and have the generating AI perform anomaly detection.

[0068] The AI ​​control system is equipped with an emergency stop function. The emergency stop function can immediately stop the system if undesirable behavior occurs. The emergency stop function stops the AI's operation, for example, by button operation or sensor detection. The emergency stop function activates automatically when the AI ​​takes abnormal behavior. The emergency stop function is equipped with software control to safely stop the AI's operation. This allows the emergency stop function to immediately stop the system if undesirable behavior occurs. Some or all of the above-described processes in the emergency stop function may be performed using AI, for example, or without AI. For example, the emergency stop function can input abnormal data detected by sensors into a generating AI and have the generating AI make a decision to perform an emergency stop.

[0069] The AI ​​control system includes a fail-safe system. The fail-safe system ensures safety even in the event of a malfunction. The fail-safe system includes, for example, redundant systems and backup functions. The fail-safe system automatically transitions to a safe state if the AI ​​malfunctions. The fail-safe system is equipped with safety devices that activate when an AI malfunction is detected. This ensures safety even in the event of a malfunction. Some or all of the above-described processes in the fail-safe system may be performed using, for example, the AI, or not using the AI. For example, the fail-safe system can have the generating AI perform the switching of the redundant system.

[0070] The listing unit can estimate the user's emotions and determine the priority of actions to list based on the estimated emotions. For example, if the user is stressed, the listing unit will prioritize actions that reduce stress among the actions to list. For example, if the user is relaxed, the listing unit may also prioritize actions that maintain relaxation among the actions to list. For example, if the user is in a hurry, the listing unit may also prioritize actions that can be completed quickly among the actions to list. This allows the priority of actions to be adjusted according to the user's emotions. Emotion estimation is achieved using an emotion estimation function, for example, using 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 listing unit may be performed using AI, for example, or without AI. For example, the listing unit can input the user's facial expression data into the generative AI and have the generative AI perform emotion estimation.

[0071] The listing unit can analyze past behavioral history and improve the accuracy of the behaviors listed. For example, the listing unit can prioritize displaying behaviors that the user has frequently performed in the past when listing them. The listing unit can also predict and list behaviors to be performed during specific time periods based on the user's past behavioral history. The listing unit can also analyze the user's past behavioral patterns and list the most efficient behaviors. This allows for improved accuracy of behavior based on past behavioral history. Some or all of the above-described processes in the listing unit may be performed using AI, for example, or without AI. For example, the listing unit can input past behavioral data into a generating AI and have the generating AI perform behavioral predictions.

[0072] The listing unit can dynamically update the actions to be listed according to specific time periods or situations. For example, the listing unit can list actions that a user performs in the morning and different actions in the afternoon and evening. The listing unit can also list actions appropriate for a specific location if the user is in that location. The listing unit can also dynamically update the actions to be listed according to the user's current situation (e.g., at work or on vacation). This allows for dynamic updates of actions according to time periods and situations. Some or all of the above processing in the listing unit may be performed using AI, for example, or without AI. For example, the listing unit can input the user's current situation data into a generating AI and have the generating AI perform the action update.

[0073] The rule setting unit can estimate the user's emotions and adjust the timing of rule application based on the estimated user emotions. For example, if the user is stressed, the rule setting unit may delay the application of the rules. For example, if the user is relaxed, the rule setting unit may also expedite the application of the rules. For example, if the user is in a hurry, the rule setting unit may also expedite the application of the rules. This allows the timing of rule application to be adjusted according to the user's emotions. Emotion estimation is achieved using an emotion estimation function, for example, using 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 rule setting unit may be performed using AI, for example, or without AI. For example, the rule setting unit can input user facial expression data into a generative AI and have the generative AI perform emotion estimation.

[0074] The rule setting unit can analyze past rule application history and improve the accuracy of rules. For example, the rule setting unit can analyze the results of rules previously applied by the user and set the optimal rule. For example, the rule setting unit can predict and adjust rule application in specific situations based on the user's past rule application history. For example, the rule setting unit can analyze the user's past rule application patterns and set the most effective rule. This allows for improved rule accuracy based on past rule application history. Some or all of the above processes in the rule setting unit may be performed using AI, for example, or without AI. For example, the rule setting unit can input past rule application data into a generating AI and have the generating AI perform rule optimization.

[0075] The rule setting unit can dynamically change rules according to specific time periods or situations. For example, the rule setting unit can set rules to apply to the user during the morning hours and different rules to apply during the day and at night. The rule setting unit can also set rules appropriate for a specific location if the user is in that location. The rule setting unit can also dynamically change rules according to the user's current situation (e.g., at work or on vacation). This allows for dynamic rule changes according to time periods and situations. Some or all of the above-described processes in the rule setting unit may be performed using AI, for example, or without AI. For example, the rule setting unit can input the user's current situation data into a generating AI and have the generating AI execute rule changes.

[0076] The reward / penalty unit can estimate the user's emotions and adjust the content of rewards and penalties based on the estimated user emotions. For example, if the user is stressed, the reward / penalty unit can provide a reward that reduces stress. For example, if the user is relaxed, the reward / penalty unit can also provide a reward that helps maintain relaxation. For example, if the user is in a hurry, the reward / penalty unit can also provide a reward that helps complete tasks quickly. This allows the content of rewards and penalties to be adjusted according to the user's emotions. Emotion estimation is achieved using an emotion estimation function, for example, using 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 reward / penalty unit may be performed using AI, for example, or not using AI. For example, the reward / penalty unit can input the user's facial expression data into a generative AI and have the generative AI perform emotion estimation.

[0077] The reward and penalty unit can analyze past reward and penalty history to improve the accuracy of rewards and penalties. For example, the reward and penalty unit can analyze the results of rewards and penalties a user has received in the past and set the optimal reward and penalty. For example, the reward and penalty unit can predict and adjust rewards and penalties in specific situations based on a user's past reward and penalty history. For example, the reward and penalty unit can analyze a user's past reward and penalty patterns and set the most effective reward and penalty. This allows for improved accuracy of rewards and penalties based on past reward and penalty history. Some or all of the above processing in the reward and penalty unit may be performed using AI, for example, or without AI. For example, the reward and penalty unit can input past reward and penalty data into a generating AI and have the generating AI perform the optimization of rewards and penalties.

[0078] The reward / penalty unit can dynamically change rewards and penalties according to specific time periods or situations. For example, the reward / penalty unit can set rewards and penalties for a user in the morning and apply different rewards and penalties during the day and night. The reward / penalty unit can also set rewards and penalties appropriate to a user's location if the user is in a specific place. The reward / penalty unit can also dynamically change rewards and penalties according to the user's current situation (e.g., at work or on vacation). This allows for dynamic changes to rewards and penalties according to time periods and situations. Some or all of the above-described processes in the reward / penalty unit may be performed using AI, for example, or without AI. For example, the reward / penalty unit can input the user's current situation data into a generating AI and have the generating AI execute the changes to rewards and penalties.

[0079] The feedback unit can estimate the user's emotions and adjust the content of the feedback based on the estimated emotions. For example, if the user is stressed, the feedback unit can provide feedback to reduce stress. For example, if the user is relaxed, the feedback unit can also provide feedback to maintain relaxation. For example, if the user is in a hurry, the feedback unit can also provide feedback to complete the task quickly. This allows the content of the feedback to be adjusted according to the user's emotions. Emotion estimation is achieved using an emotion estimation function, for example, using 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 feedback unit may be performed using AI, for example, or without AI. For example, the feedback unit can input the user's facial expression data into the generative AI and have the generative AI perform emotion estimation.

[0080] The feedback unit can analyze past feedback history and improve the accuracy of feedback. For example, the feedback unit can analyze the results of feedback the user has received in the past and set the optimal feedback. The feedback unit can also predict and adjust feedback in specific situations based on the user's past feedback history. For example, the feedback unit can analyze the user's past feedback patterns and set the most effective feedback. This allows for improved feedback accuracy based on past feedback history. Some or all of the above processes in the feedback unit may be performed using AI, for example, or without AI. For example, the feedback unit can input past feedback data into a generating AI and have the generating AI perform feedback optimization.

[0081] The feedback unit can dynamically change the feedback depending on the time of day or situation. For example, the feedback unit can set the feedback a user receives in the morning and apply different feedback during the day and at night. For example, the feedback unit can also set feedback appropriate for a specific location if the user is in a particular place. For example, the feedback unit can dynamically change the feedback depending on the user's current situation (e.g., working or on vacation). This allows for dynamic changes to the feedback depending on the time of day or situation. Some or all of the above processing in the feedback unit may be performed using AI, for example, or without AI. For example, the feedback unit can input the user's current situation data into a generating AI and have the generating AI perform the feedback changes.

[0082] The ethical judgment unit can estimate the user's emotions and adjust the criteria for ethical judgments based on the estimated user emotions. For example, if the user is stressed, the ethical judgment unit will prioritize ethical judgments that reduce stress. For example, if the user is relaxed, the ethical judgment unit may also prioritize ethical judgments that maintain relaxation. For example, if the user is in a hurry, the ethical judgment unit may also prioritize ethical judgments that allow for quick completion. This allows the criteria for ethical judgments to be adjusted according to the user's emotions. Emotion estimation is achieved using an emotion estimation function, for example, using 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 ethical judgment unit may be performed using AI, for example, or without AI. For example, the ethical judgment unit can input user facial expression data into a generative AI and have the generative AI perform emotion estimation.

[0083] The ethical judgment unit can analyze past ethical judgment history and improve the accuracy of ethical judgments. For example, the ethical judgment unit can analyze the results of ethical judgments made by the user in the past and set the optimal ethical judgment. For example, the ethical judgment unit can predict and adjust ethical judgments in specific situations based on the user's past ethical judgment history. For example, the ethical judgment unit can analyze the user's past ethical judgment patterns and set the most effective ethical judgment. This allows for improved accuracy of ethical judgments based on past ethical judgment history. Some or all of the above processes in the ethical judgment unit may be performed using AI, for example, or without AI. For example, the ethical judgment unit can input past ethical judgment data into a generating AI and have the generating AI perform the optimization of ethical judgments.

[0084] The ethical judgment unit can dynamically change ethical judgments according to specific time periods or circumstances. For example, the ethical judgment unit can set ethical judgments for the user in the morning and apply different ethical judgments during the day and at night. For example, the ethical judgment unit can also set ethical judgments appropriate to a specific location if the user is in that location. For example, the ethical judgment unit can dynamically change ethical judgments according to the user's current situation (e.g., at work or on vacation). This allows for dynamic changes to ethical judgments depending on the time period and circumstances. Some or all of the above processing in the ethical judgment unit may be performed using AI, for example, or without AI. For example, the ethical judgment unit can input the user's current situation data into a generating AI and have the generating AI perform the change in ethical judgments.

[0085] The monitoring system can estimate the user's emotions and adjust the monitoring frequency based on the estimated emotions. For example, if the user is stressed, the monitoring system may lower the monitoring frequency. For example, if the user is relaxed, the monitoring system may also increase the monitoring frequency. For example, if the user is in a hurry, the monitoring system may also increase the monitoring frequency. This allows the monitoring frequency to be adjusted according to the user's emotions. Emotion estimation is achieved using an emotion estimation function, for example, using 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 monitoring system may be performed using AI, for example, or not using AI. For example, the monitoring system can input user facial expression data into a generative AI and have the generative AI perform emotion estimation.

[0086] The monitoring system can analyze past monitoring history to improve monitoring accuracy. For example, the monitoring system can analyze the results of past monitoring received by the user and set optimal monitoring. For example, the monitoring system can predict and adjust monitoring in specific situations based on the user's past monitoring history. For example, the monitoring system can analyze the user's past monitoring patterns and set the most effective monitoring. This allows for improved monitoring accuracy based on past monitoring history. Some or all of the above processes in the monitoring system may be performed using AI, for example, or without AI. For example, the monitoring system can input past monitoring data into a generating AI and have the generating AI perform monitoring optimization.

[0087] The emergency stop function can estimate the user's emotions and adjust the criteria for emergency stop based on the estimated emotions. For example, if the user is stressed, the emergency stop function may relax the criteria for emergency stop. For example, if the user is relaxed, the emergency stop function may tighten the criteria for emergency stop. For example, if the user is in a hurry, the emergency stop function may perform an emergency stop quickly. This allows the criteria for emergency stop to be adjusted according to the user's emotions. Emotion estimation is achieved using an emotion estimation function, for example, using 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 emergency stop function may be performed using AI or not using AI. For example, the emergency stop function can input user facial expression data into a generative AI and have the generative AI perform emotion estimation.

[0088] The emergency stop function can analyze past emergency stop history to improve the accuracy of emergency stops. For example, the emergency stop function can analyze the results of emergency stops performed by the user in the past and set the optimal emergency stop. For example, the emergency stop function can predict and adjust emergency stops in specific situations based on the user's past emergency stop history. For example, the emergency stop function can analyze the user's past emergency stop patterns and set the most effective emergency stop. This allows for improved accuracy of emergency stops based on past emergency stop history. Some or all of the above processes in the emergency stop function may be performed using AI, for example, or without AI. For example, the emergency stop function can input past emergency stop data into a generating AI and have the generating AI perform emergency stop optimization.

[0089] A fail-safe system can estimate the user's emotions and adjust the fail-safe criteria based on the estimated emotions. For example, if the user is stressed, the fail-safe system may relax the fail-safe criteria. For example, if the user is relaxed, the fail-safe system may tighten the fail-safe criteria. For example, if the user is in a hurry, the fail-safe system may perform a fail-safe quickly. This allows the fail-safe criteria to be adjusted according to the user's emotions. Emotion estimation is achieved using an emotion estimation function, for example, using 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 fail-safe system may be performed using AI or not using AI. For example, the fail-safe system can input user facial expression data into a generative AI and have the generative AI perform emotion estimation.

[0090] A fail-safe system can analyze past fail-safe history to improve the accuracy of its fail-safes. For example, a fail-safe system can analyze the results of past fail-safes performed by a user and set the optimal fail-safe. A fail-safe system can also predict and adjust fail-safes in specific situations based on a user's past fail-safe history. A fail-safe system can also analyze a user's past fail-safe patterns and set the most effective fail-safe. This allows for improved fail-safe accuracy based on past fail-safe history. Some or all of the above processes in a fail-safe system may be performed using AI, for example, or without AI. For example, a fail-safe system can input past fail-safe data into a generating AI and have the generating AI perform fail-safe optimization.

[0091] A fail-safe system can estimate a user's emotions and determine the priority of fail-safes based on the estimated emotions. For example, if the user is stressed, the fail-safe system might prioritize fail-safes to reduce stress. If the user is relaxed, the fail-safe system might also prioritize fail-safes to maintain relaxation. If the user is in a hurry, the fail-safe system might also prioritize fail-safes to complete tasks quickly. This allows the system to determine the priority of fail-safes according to the user's emotions. Emotion estimation is achieved using an emotion estimation function, for example, with 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-described processes in the fail-safe system may be performed using AI or not. For example, the fail-safe system can input user facial expression data into a generative AI and have the generative AI perform emotion estimation.

[0092] The failsafe system can customize its failsafes based on the user's geographical location. For example, if the user is in a specific city, the failsafe system can set up failsafes to apply in that city. For example, if the user is traveling, the failsafe system can also set up failsafes to apply at the travel destination. For example, if the user is at home, the failsafe system can also set up failsafes to apply at home. This allows the failsafe to be customized based on the user's geographical location. Some or all of the above processes in the failsafe system may be performed using AI, for example, or not using AI. For example, the failsafe system can input the user's geographical location data into a generating AI and have the generating AI perform the failsafe customization.

[0093] The system according to the embodiment is not limited to the example described above, and various modifications are possible, for example, as follows.

[0094] The AI ​​control system can estimate the user's emotions and dynamically change the priority of actions based on the estimated emotions. For example, if the user is stressed, actions that reduce stress will be prioritized. If the user is relaxed, actions that maintain relaxation may be prioritized. If the user is in a hurry, actions that can be completed quickly may be prioritized. This allows the priority of actions to be adjusted according to the user's emotions. Emotion estimation is achieved using an emotion engine or a generative AI, etc. The generative AI is, but is not limited to, text generation AI or multimodal generation AI. Some or all of the above processing in the listing unit may be performed using AI or not using AI. For example, the listing unit can input the user's facial expression data into the generative AI and have the generative AI perform emotion estimation.

[0095] The AI ​​control system can estimate the user's emotions and adjust the timing of rule application based on the estimated emotions. For example, if the user is stressed, the rule application can be delayed. If the user is relaxed, the rule application can be accelerated. If the user is in a hurry, the rule application can be performed quickly. This allows the timing of rule application to be adjusted according to the user's emotions. Emotion estimation is achieved using an emotion engine or a generative AI, etc. The generative AI is, but is not limited to, text generation AI or multimodal generation AI. Some or all of the above processing in the rule setting unit may be performed using AI or not. For example, the rule setting unit can input user facial expression data into the generative AI and have the generative AI perform emotion estimation.

[0096] The AI ​​control system can estimate the user's emotions and adjust the content of rewards and penalties based on the estimated user emotions. For example, if the user is stressed, it can provide a reward that reduces stress. If the user is relaxed, it can also provide a reward that helps maintain that relaxation. If the user is in a hurry, it can also provide a reward that helps them complete tasks quickly. This allows the content of rewards and penalties to be adjusted according to the user's emotions. Emotion estimation is achieved using an emotion engine or a generative AI, etc. The generative AI is, but is not limited to, text generation AI or multimodal generation AI. Some or all of the above processing in the reward / penalty unit may be performed using AI or not. For example, the reward / penalty unit can input the user's facial expression data into the generative AI and have the generative AI perform emotion estimation.

[0097] The AI ​​control system can estimate the user's emotions and adjust the content of the feedback based on the estimated emotions. For example, if the user is stressed, it can provide feedback to reduce stress. If the user is relaxed, it can provide feedback to maintain that relaxation. If the user is in a hurry, it can provide feedback to complete the task quickly. This allows the content of the feedback to be adjusted according to the user's emotions. Emotion estimation is achieved using an emotion engine or a generative AI, etc. The generative AI is, but is not limited to, text generation AI or multimodal generation AI. Some or all of the above processing in the feedback unit may be performed using AI or not. For example, the feedback unit can input the user's facial expression data into a generative AI and have the generative AI perform emotion estimation.

[0098] The AI ​​control system can estimate the user's emotions and adjust the monitoring frequency based on the estimated emotions. For example, if the user is stressed, the monitoring frequency can be lowered. If the user is relaxed, the monitoring frequency can be increased. If the user is in a hurry, the monitoring frequency can be increased rapidly. This allows the monitoring frequency to be adjusted according to the user's emotions. Emotion estimation is achieved using an emotion engine or generative AI, etc. Generative AI is, but is not limited to, text generation AI or multimodal generation AI. Some or all of the above processing in the monitoring system may be performed using AI or not. For example, the monitoring system can input user facial expression data into a generative AI and have the generative AI perform emotion estimation.

[0099] The AI ​​control system can analyze past behavioral history and improve the accuracy of the actions listed. For example, it can prioritize displaying actions that a user has frequently performed in the past when listing them. It can also predict and list actions that a user will perform during specific time periods based on their past behavioral history. It can also analyze a user's past behavioral patterns and list the most efficient actions. This allows for improved accuracy of actions based on past behavioral history. Some or all of the above processing in the listing unit may be performed using AI or not. For example, the listing unit can input past behavioral data into a generating AI and have the generating AI perform behavioral predictions.

[0100] The AI ​​control system can dynamically update the list of actions according to specific time periods and situations. For example, it can list actions a user takes in the morning and different actions for the afternoon and evening. If the user is in a specific location, it can also list actions appropriate for that location. It can also dynamically update the list of actions according to the user's current situation (e.g., at work or on vacation). This allows for dynamic updates of actions according to time periods and situations. Some or all of the above processing in the listing unit may be performed using AI or not. For example, the listing unit can input the user's current situation data into a generating AI and have the generating AI perform the action update.

[0101] The AI ​​control system can analyze past rule application history and improve the accuracy of rules. For example, it can analyze the results of rules previously applied by the user and set the optimal rules. It can also predict and adjust rule application in specific situations based on the user's past rule application history. It can also analyze the user's past rule application patterns and set the most effective rules. This allows for improved rule accuracy based on past rule application history. Some or all of the above processing in the rule setting unit may be performed using AI or not. For example, the rule setting unit can input past rule application data into a generating AI and have the generating AI perform rule optimization.

[0102] The AI ​​control system can dynamically change rules according to specific time periods or situations. For example, it can set rules to apply to the user in the morning and different rules to apply during the day and at night. If the user is in a specific location, it can also set rules appropriate for that location. It can also dynamically change rules according to the user's current situation (e.g., working or on vacation). This allows for dynamic rule changes according to time periods and situations. Some or all of the above-described processes in the rule setting unit may be performed using AI or not. For example, the rule setting unit can input user's current situation data into a generating AI and have the generating AI execute rule changes.

[0103] The AI ​​control system can analyze past reward and penalty history to improve the accuracy of rewards and penalties. For example, it can analyze the results of rewards and penalties a user has received in the past and set the optimal rewards and penalties. It can also predict and adjust rewards and penalties in specific situations based on the user's past reward and penalty history. It can also analyze the user's past reward and penalty patterns and set the most effective rewards and penalties. This allows for improved accuracy of rewards and penalties based on past reward and penalty history. Some or all of the above processing in the reward and penalty unit may be performed using AI or not. For example, the reward and penalty unit can input past reward and penalty data into a generating AI and have the generating AI perform the optimization of rewards and penalties.

[0104] The following briefly describes the processing flow for example form 2.

[0105] Step 1: The listing unit lists permitted and prohibited actions. For example, it sets up a whitelist / blacklist for the AI ​​and lists permitted and prohibited actions. The listing unit pre-defines the actions the AI ​​will perform and controls it to execute only permitted actions. Step 2: The rule setting unit sets actions that are permitted only under specific circumstances. For example, it restricts actions based on specific time periods, locations, or environmental conditions. The rule setting unit controls the AI ​​to perform actions only when specific conditions are met. Step 3: The reward-penalty unit rewards desirable behaviors and penalizes undesirable behaviors. For example, reinforcement learning is used to evaluate the AI's behavior, rewarding desirable behaviors and penalizing undesirable behaviors. The reward-penalty unit controls the AI ​​so that it reinforces desirable behaviors and suppresses undesirable behaviors through learning. Step 4: The feedback unit adjusts the algorithm based on the AI's actions. For example, it monitors the AI's actions and adjusts the algorithm's parameters. The feedback unit optimizes the algorithm so that the AI ​​takes appropriate actions. Step 5: The Ethics Judgment Unit makes ethical judgments. For example, it evaluates whether the actions taken by the AI ​​comply with ethical guidelines. The Ethics Judgment Unit controls the AI ​​to act ethically.

[0106] 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.

[0107] 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.

[0108] 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.

[0109] Each of the multiple elements described above, including the listing unit, rule setting unit, reward / penalty unit, feedback unit, and ethical judgment unit, is implemented in at least one of the smart device 14 and the data processing unit 12. For example, the listing unit is implemented by the control unit 46A of the smart device 14, which sets up whitelists / blacklists and lists permitted and prohibited actions. The rule setting unit is implemented by the specific processing unit 290 of the data processing unit 12, which sets actions that are permitted only under specific circumstances. The reward / penalty unit is implemented by the control unit 46A of the smart device 14, which rewards desirable actions and penalizes undesirable actions. The feedback unit is implemented by the specific processing unit 290 of the data processing unit 12, which adjusts the algorithm based on the AI's actions. The ethical judgment unit is implemented by the control unit 46A of the smart device 14, which evaluates whether the actions taken by the AI ​​comply with ethical guidelines. The correspondence between each unit and the device or control unit is not limited to the example described above, and various modifications are possible.

[0110] [Second Embodiment] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0111] 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.

[0112] 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.

[0113] 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.

[0114] The microphone 238 receives voice signals from the user and accepts instructions from the user. 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.

[0115] 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).

[0116] 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.

[0117] 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.

[0118] 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.

[0119] 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.

[0120] 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.

[0121] 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.).

[0122] 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.

[0123] 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.

[0124] 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.

[0125] Each of the multiple elements described above, including the listing unit, rule setting unit, reward / penalty unit, feedback unit, and ethical judgment unit, is implemented in at least one of the smart glasses 214 and the data processing unit 12. For example, the listing unit is implemented by the control unit 46A of the smart glasses 214, which sets up whitelists / blacklists and lists permitted and prohibited actions. The rule setting unit is implemented by the specific processing unit 290 of the data processing unit 12, which sets actions that are permitted only under specific circumstances. The reward / penalty unit is implemented by the control unit 46A of the smart glasses 214, which rewards desirable actions and penalizes undesirable actions. The feedback unit is implemented by the specific processing unit 290 of the data processing unit 12, which adjusts the algorithm based on the AI's actions. The ethical judgment unit is implemented by the control unit 46A of the smart glasses 214, which evaluates whether the actions taken by the AI ​​comply with ethical guidelines. The correspondence between each unit and the device or control unit is not limited to the example described above, and various modifications are possible.

[0126] [Third Embodiment] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0127] 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.

[0128] 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.

[0129] 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.

[0130] The microphone 238 receives voice signals from the user and accepts instructions from the user. 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.

[0131] 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).

[0132] 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.

[0133] 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.

[0134] 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.

[0135] 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.

[0136] 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.

[0137] 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.).

[0138] 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.

[0139] 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.

[0140] 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.

[0141] Each of the multiple elements described above, including the listing unit, rule setting unit, reward / penalty unit, feedback unit, and ethical judgment unit, is implemented in at least one of the headset terminal 314 and the data processing unit 12. For example, the listing unit is implemented by the control unit 46A of the headset terminal 314, setting whitelists / blacklists and listing permitted and prohibited actions. The rule setting unit is implemented by the specific processing unit 290 of the data processing unit 12, setting actions that are permitted only under specific circumstances. The reward / penalty unit is implemented by the control unit 46A of the headset terminal 314, for example, rewarding desirable actions and imposing penalties on undesirable actions. The feedback unit is implemented by the specific processing unit 290 of the data processing unit 12, for example, adjusting the algorithm based on the AI's actions. The ethical judgment unit is implemented by the control unit 46A of the headset terminal 314, for example, evaluating whether the actions taken by the AI ​​comply with ethical guidelines. The correspondence between each unit and the device or control unit is not limited to the example described above, and various modifications are possible.

[0142] [Fourth Embodiment] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[0143] 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.

[0144] 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.

[0145] 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.

[0146] The microphone 238 receives voice signals from the user and accepts instructions from the user. 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.

[0147] 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).

[0148] 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.

[0149] 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.

[0150] 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.

[0151] 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.

[0152] 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.

[0153] 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.

[0154] 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.).

[0155] 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.

[0156] 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.

[0157] 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.

[0158] Each of the multiple elements described above, including the listing unit, rule setting unit, reward / penalty unit, feedback unit, and ethical judgment unit, is implemented in at least one of the robot 414 and the data processing unit 12. For example, the listing unit is implemented by the control unit 46A of the robot 414, setting whitelists / blacklists and listing permitted and prohibited actions. The rule setting unit is implemented by the specific processing unit 290 of the data processing unit 12, setting actions that are permitted only under specific circumstances. The reward / penalty unit is implemented by the control unit 46A of the robot 414, for example, rewarding desirable actions and imposing penalties on undesirable actions. The feedback unit is implemented by the specific processing unit 290 of the data processing unit 12, for example, adjusting the algorithm based on the AI's actions. The ethical judgment unit is implemented by the control unit 46A of the robot 414, for example, evaluating whether the actions taken by the AI ​​comply with ethical guidelines. The correspondence between each unit and the device or control unit is not limited to the examples described above and can be modified in various ways.

[0159] 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.

[0160] 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.

[0161] 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.

[0162] 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.

[0163] 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.

[0164] 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."

[0165] 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.

[0166] 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.

[0167] 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.

[0168] 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.

[0169] 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.

[0170] 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.

[0171] 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.

[0172] 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.

[0173] 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.

[0174] 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.

[0175] 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.

[0176] 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.

[0177] (Note 1) A listing unit that lists permitted and prohibited actions, A rule setting section that sets actions permitted only under specific circumstances, A reward-penalty unit that rewards desirable behavior and penalizes undesirable behavior, A feedback unit that adjusts the algorithm based on the AI's actions, It comprises an ethical judgment unit that makes ethical judgments, A system characterized by the following features. (Note 2) Equipped with a monitoring system The system described in Appendix 1, characterized by the features described herein. (Note 3) Equipped with an emergency stop function. The system described in Appendix 1, characterized by the features described herein. (Note 4) Equipped with a fail-safe system The system described in Appendix 1, characterized by the features described herein. (Note 5) The listing unit, It estimates the user's emotions and determines the priority of actions to list based on those estimated emotions. The system described in Appendix 1, characterized by the features described herein. (Note 6) The listing unit, Analyze past behavioral history to improve the accuracy of the behaviors listed. The system described in Appendix 1, characterized by the features described herein. (Note 7) The listing unit, The list of actions is dynamically updated according to specific time periods and circumstances. The system described in Appendix 1, characterized by the features described herein. (Note 8) The aforementioned rule setting unit, It estimates the user's emotions and adjusts the timing of rule application based on the estimated user emotions. The system described in Appendix 1, characterized by the features described herein. (Note 9) The aforementioned rule setting unit, Analyze past rule application history to improve rule accuracy. The system described in Appendix 1, characterized by the features described herein. (Note 10) The aforementioned rule setting unit, The rules are dynamically changed according to specific time periods or circumstances. The system described in Appendix 1, characterized by the features described herein. (Note 11) The aforementioned reward penalty section is, The system estimates the user's emotions and adjusts rewards and penalties based on those estimated emotions. The system described in Appendix 1, characterized by the features described herein. (Note 12) The aforementioned reward penalty section is, We analyze past reward and penalty history to improve the accuracy of rewards and penalties. The system described in Appendix 1, characterized by the features described herein. (Note 13) The aforementioned reward penalty section is, Dynamically change rewards and penalties based on specific time periods or circumstances. The system described in Appendix 1, characterized by the features described herein. (Note 14) The aforementioned feedback unit is It estimates the user's emotions and adjusts the content of the feedback based on those estimated emotions. The system described in Appendix 1, characterized by the features described herein. (Note 15) The aforementioned feedback unit is Analyze past feedback history to improve the accuracy of feedback. The system described in Appendix 1, characterized by the features described herein. (Note 16) The aforementioned feedback unit is Dynamically change the feedback based on specific time periods or circumstances. The system described in Appendix 1, characterized by the features described herein. (Note 17) The aforementioned ethics judgment department, We estimate user sentiment and adjust ethical judgment standards based on the estimated user sentiment. The system described in Appendix 1, characterized by the features described herein. (Note 18) The aforementioned ethics judgment department, Analyzing past ethical judgment history improves the accuracy of ethical decisions. The system described in Appendix 1, characterized by the features described herein. (Note 19) The aforementioned ethics judgment department, Ethical judgments are dynamically changed depending on specific time periods and circumstances. The system described in Appendix 1, characterized by the features described herein. (Note 20) The aforementioned monitoring system It estimates the user's emotions and adjusts the monitoring frequency based on the estimated emotions. The system described in Appendix 2, characterized by the features described herein. (Note 21) The aforementioned monitoring system Analyze past monitoring history to improve monitoring accuracy. The system described in Appendix 2, characterized by the features described herein. (Note 22) The aforementioned emergency stop function, The system estimates the user's emotions and adjusts the emergency stop criteria based on those emotions. The system described in Appendix 3, characterized by the features described herein. (Note 23) The aforementioned emergency stop function, Analyze past emergency stop history to improve the accuracy of emergency stops. The system described in Appendix 3, characterized by the features described herein. (Note 24) The aforementioned fail-safe system is Estimate user sentiment and adjust fail-safe criteria based on the estimated user sentiment. The system described in Appendix 4, characterized by the features described herein. (Note 25) The aforementioned fail-safe system is Analyze past fail-safe history to improve the accuracy of fail-safes. The system described in Appendix 4, characterized by the features described herein. (Note 26) The aforementioned fail-safe system is It estimates the user's emotions and determines fail-safe priorities based on those estimated emotions. The system described in Appendix 4, characterized by the features described herein. (Note 27) The aforementioned fail-safe system is Customize fail-safe mechanisms based on the user's geographical location. The system described in Appendix 4, characterized by the features described herein. [Explanation of symbols]

[0178] 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. A listing unit that lists permitted and prohibited actions, A rule setting section that sets actions permitted only under specific circumstances, A reward-penalty unit that rewards desirable behavior and penalizes undesirable behavior, A feedback unit that adjusts the algorithm based on the AI's actions, It comprises an ethical judgment unit that makes ethical judgments, A system characterized by the following features.

2. Equipped with a monitoring system The system according to feature 1.

3. Equipped with an emergency stop function. The system according to feature 1.

4. Equipped with a fail-safe system The system according to feature 1.

5. The listing unit, It estimates the user's emotions and determines the priority of actions to list based on those estimated emotions. The system according to feature 1.

6. The listing unit, Analyze past behavioral history to improve the accuracy of the behaviors listed. The system according to feature 1.

7. The listing unit, The list of actions is dynamically updated according to specific time periods and circumstances. The system according to feature 1.

8. The aforementioned rule setting unit, It estimates the user's emotions and adjusts the timing of rule application based on the estimated user emotions. The system according to feature 1.

9. The aforementioned rule setting unit, Analyze past rule application history to improve rule accuracy. The system according to feature 1.

10. The aforementioned rule setting unit, The rules are dynamically changed according to specific time periods or circumstances. The system according to feature 1.

11. The aforementioned reward penalty section is, The system estimates the user's emotions and adjusts rewards and penalties based on those estimated emotions. The system according to feature 1.

Images