System

Abstract

To provide a system that effectively searches for vulnerabilities in the system and generates and evaluates defensive measures.SOLUTION: A system according to the embodiment comprises an attack unit, a defense unit, and a feedback unit. The attack unit searches for vulnerabilities in the system. The defense unit generates defensive measures based on an attack scenario generated by the attack unit. The feedback unit evaluates the effectiveness of the defensive measures generated by the defense unit, and generates a new attack scenario.SELECTED DRAWING: Figure 1

Description

[Technical field]

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

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

[0003] [Patent Document 1] JP 2022-180282 A Summary of the Invention [Problem to be solved by the invention]

[0004] In conventional techniques, the process of effectively searching for system vulnerabilities and generating defense measures is complicated and there is room for improvement.

[0005] The system according to the embodiment aims to effectively search for vulnerabilities in a system and generate and evaluate defensive measures. [Means for solving the problem]

[0006] A system according to an embodiment includes an attack unit, a defense unit, and a feedback unit. The attack unit searches for vulnerabilities in the system. The defense unit generates defense measures based on the attack scenario generated by the attack unit. The feedback unit evaluates the effectiveness of the defense measures generated by the defense unit and generates a new attack scenario. Effect of the Invention

[0007] The system according to the embodiment can effectively search for vulnerabilities in a system and generate and evaluate defensive measures. [Brief description of the drawings]

[0008] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Diagram 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. FIG. [Diagram 3] FIG. 11 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Diagram 5] FIG. 13 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 13 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 13 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0009] Hereinafter, an example of an embodiment of a system according to the technology of the present disclosure will be described with reference to the accompanying drawings.

[0010] First, the terms used in the following description will be explained.

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

[0012] In the following embodiments, a signed RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by the processor.

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

[0014] In the following embodiments, a communication I / F (Interface) with a code is an interface including a communication processor and an antenna. The communication I / F controls communication between multiple computers. An example of a communication standard applied to the communication I / F is a wireless communication standard including 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." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. In addition, in this specification, the same idea as "A and / or B" is also applied when three or more things are expressed by connecting them with "and / or."

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

[0017] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[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. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

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

[0020] The reception device 38 includes a touch panel 38A and a microphone 38B, and receives user input. The touch panel 38A detects contact with a pointer (e.g., a pen or a finger) to receive user input by the touch of the pointer. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 (see FIG. 2) acquires the data indicating the user input.

[0021] Output device 40 includes a display 40A and a speaker 40B, and presents data to a user by outputting the data in a form of expression that the user can perceive (e.g., voice and / or text). Display 40A displays visible information such as text and images according to instructions from processor 46. Speaker 40B outputs voice according to instructions from processor 46. Camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a Complementary Metal-Oxide-Semiconductor (CMOS) image sensor or a Charge Coupled Device (CCD) image sensor.

[0022] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for transmitting and receiving various types of information between the processor 46 and the processor 28 via the network 54.

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

[0024] As shown in Fig. 2, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32. The specific process program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific process program 56 from the storage 32, and executes the read specific process program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific process program 56 executed on the RAM 30.

[0025] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290. The identification processing unit 290 can estimate the emotion of the user using the emotion identification model 59 and perform identification processing using the user's emotion. In the emotion estimation function (emotion identification function) using the emotion identification model 59, various estimations and predictions regarding the user's emotion, including estimation and prediction of the user's emotion, are performed, but are not limited to such examples. In addition, the estimation and prediction of the emotion also include, for example, analysis of the emotion.

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

[0027] Note that a device other than the data processing device 12 may 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 a processing result (prediction result, etc.) using the data generation model 58 by communicating with the server device having the data generation model 58. Also, the data processing device 12 may be a server device, or may be a terminal device (e.g., a mobile phone, a robot, a home appliance, etc.) owned by a user. Next, an example of processing by the data processing system 10 according to the first embodiment will be described.

[0028] (Example 1) A penetration test support system according to an embodiment of the present invention is a system that supports penetration tests using a generation AI that has been trained in two patterns, one for an attack team and one for a defense team. In this system, the generation AI for the attack team searches for vulnerabilities in the system and generates an attack scenario. Next, the generation AI for the defense team generates a defense measure against the attack scenario. This makes it possible to strengthen the security of the system from both the attack and defense perspectives. For example, the generation AI for the attack team analyzes each component of the system in detail and identifies potential vulnerabilities. For example, unupdated software and improper settings are examples of vulnerabilities. Next, the generation AI generates an attack scenario based on the identified vulnerability. For example, attack methods such as SQL injection and cross-site scripting (XSS) are included. Next, the generation AI for the defense team generates a defense measure against the attack scenario. The generation AI analyzes the attack scenario and proposes an optimal defense measure. For example, firewall setting changes and software updates are examples of defense measures. Furthermore, the generation AI for the defense team simulates the effect of the proposed defense measure and evaluates the actual effect. Finally, the generation AI for the attack team and the generation AI for the defense team provide feedback to each other. The generation AI for the attack team evaluates the effectiveness of defensive measures and generates new attack scenarios. Meanwhile, the generation AI for the defense team again generates defensive measures against new attack scenarios. By repeating this process, the security of the system can be continuously improved. This makes it possible to strengthen the security of the system from both sides of attack and defense. For example, if the generation AI for the attack team discovers a new vulnerability, the generation AI for the defense team can immediately propose countermeasures, quickly improving the security of the system. In addition, through a feedback loop, attack and defense skills are improved, enabling more advanced security measures. This allows the penetration test support system to strengthen the security of the system from both sides of attack and defense.

[0029] A penetration test support system according to an embodiment includes an attack unit, a defense unit, and a feedback unit. The attack unit searches for vulnerabilities in the system. For example, the attack unit analyzes each component of the system in detail to identify potential vulnerabilities. For example, the attack unit can identify unupdated software, improper settings, and the like as vulnerabilities. The attack unit also generates an attack scenario based on the identified vulnerabilities. For example, the attack unit can generate an attack scenario including an attack method such as SQL injection or cross-site scripting (XSS). The defense unit generates a defense measure based on the attack scenario. The defense unit, for example, analyzes the attack scenario and proposes an optimal defense measure. For example, the defense unit can propose a firewall setting change, software update, and the like as a defense measure. The defense unit also simulates the effect of the proposed defense measure and evaluates the actual effect. For example, the defense unit can test the effect of the defense measure in a virtual environment and evaluate the result. The feedback unit evaluates the effect of the defense measure and generates a new attack scenario. The feedback unit can evaluate the effect of the defense measure and generate a new attack scenario based on the result. As a result, the penetration test support system according to the embodiment can strengthen the security of the system from both the attack and defense perspectives. For example, the attack unit searches for vulnerabilities in the system and generates attack scenarios. The defense unit generates defensive measures based on the attack scenarios and evaluates their effectiveness. The feedback unit evaluates the effectiveness of the defensive measures and generates new attack scenarios. As a result, the penetration test support system can continuously improve the security of the system.

[0030] The attack unit searches for vulnerabilities in the system. For example, the attack unit analyzes each component of the system in detail to identify potential vulnerabilities. Specifically, the attack unit analyzes the source code of the system using a static analysis tool to detect vulnerabilities in the code. In addition, the attack unit can use a dynamic analysis tool to monitor the behavior of the running system and detect abnormal behavior and unauthorized access. Furthermore, the attack unit uses a network scanner to analyze the network configuration of the system and identify open ports and services. This makes it possible to identify unupdated software and improper settings as vulnerabilities. In addition, the attack unit generates attack scenarios based on the identified vulnerabilities. For example, the attack unit can generate attack scenarios that include attack methods such as SQL injection and cross-site scripting (XSS). These attack scenarios are used to simulate actual attacks and are an important means for evaluating the defensive power of the system. The attack unit executes the generated attack scenarios and evaluates the impact by actually attacking the vulnerabilities of the system. This allows the attack unit to grasp the vulnerabilities of the system in detail and verify the vulnerabilities using specific attack methods.

[0031] The defense unit generates a defense measure based on the attack scenario. For example, the defense unit analyzes the attack scenario and proposes an optimal defense measure. Specifically, the defense unit analyzes the contents of the attack scenario in detail and identifies the attack method and attack path. This makes it possible to clarify the source of the attack and the purpose of the attack, and to formulate optimal defense measures against them. For example, the defense unit can propose a firewall setting change or software update as a defense measure. By changing the firewall setting, it is possible to restrict access from a specific port or IP address and prevent unauthorized access. In addition, by updating the software, it is possible to fix known vulnerabilities and strengthen the security of the system. Furthermore, the defense unit simulates the effect of the proposed defense measure and evaluates the actual effect. For example, the defense unit can test the effect of the defense measure in a virtual environment and evaluate the result. In the virtual environment, it is possible to verify the effect of the defense measure without affecting the actual system. This allows the defense unit to check whether the proposed defense measure is actually effective and to modify the defense measure as necessary. The defense unit can continuously review the defense measures and provide optimal defense measures in order to strengthen the security of the system.

[0032] The feedback unit evaluates the effectiveness of the defensive measures and generates a new attack scenario. For example, the feedback unit can evaluate the effectiveness of the defensive measures and generate a new attack scenario based on the evaluation result. Specifically, the feedback unit monitors the security state of the system after the defensive measures are implemented and evaluates how effective the defensive measures are. For example, the feedback unit analyzes the results of an attack simulation performed after the defensive measures are implemented and evaluates how much the defensive measures have prevented attacks. This makes it possible to confirm the effectiveness of the defensive measures and to modify the defensive measures as necessary. Furthermore, the feedback unit generates a new attack scenario based on the evaluation result of the effectiveness of the defensive measures. For example, if the defensive measures are effective against a specific attack method, a new attack scenario that avoids the attack method can be generated. This makes it possible to continuously strengthen the security of the system. Furthermore, the feedback unit can accumulate past attack scenarios and the effects of defensive measures in a database and utilize them for generating future attack scenarios and formulating defensive measures. This makes it possible for the feedback unit to continuously improve the security of the system and strengthen the security of the system from both attack and defense perspectives.

[0033] The attacking part can analyze each component of the system in detail to identify potential vulnerabilities. It is necessary to clarify the specific method and criteria for the detailed analysis. For example, the attacking part can perform a code review to identify vulnerabilities in the source code. The attacking part can also perform a penetration test to identify vulnerabilities in the system. For example, the attacking part can analyze each component of the system in detail to identify unfixed bugs and known security holes. As a result, potential vulnerabilities can be identified by analyzing each component of the system in detail. A part or all of the above-mentioned processing in the attacking part may be performed using, for example, a generation AI, or may be performed without using a generation AI. For example, the attacking part can input data of each component of the system into the generation AI and have the generation AI execute the identification of vulnerabilities.

[0034] The attack unit can generate an attack scenario based on the identified vulnerability. It is necessary to clarify the specific type of the identified vulnerability and the method of identification. For example, the attack unit can identify vulnerabilities such as SQL injection and cross-site scripting (XSS) and generate an attack scenario based on the vulnerability. It is necessary to clarify the specific content of the attack scenario and the method of generation. For example, the attack unit can generate an SQL injection attack scenario based on the identified vulnerability. The attack unit can also generate a cross-site scripting (XSS) attack scenario. As a result, an effective attack scenario can be created by generating an attack scenario based on the identified vulnerability. A part or all of the above-mentioned processing in the attack unit may be performed using, for example, a generation AI, or may be performed without using the generation AI. For example, the attack unit can input data of the identified vulnerability to the generation AI and cause the generation AI to generate an attack scenario.

[0035] The defense unit can analyze the attack scenario and propose appropriate defense measures. It is necessary to clarify the specific content of the appropriate defense measures and the method of proposing them. For example, the defense unit can analyze the attack scenario and propose a change in firewall settings. The defense unit can also propose software updates. In this way, the optimal defense measures can be proposed by analyzing the attack scenario. Some or all of the above-mentioned processing in the defense unit may be performed, for example, using the generation AI, or may be performed without using the generation AI. For example, the defense unit can input data of the attack scenario to the generation AI and have the generation AI execute the proposal of the defense measures.

[0036] The defense unit can simulate the effect of the proposed defense measures and evaluate the actual effect. It is necessary to clarify the specific method and criteria for simulating the effect. For example, the defense unit can test the effect of the defense measures in a virtual environment and evaluate the result. The defense unit can also perform a scenario-based simulation and evaluate the effect of the defense measures. In this way, the actual effect can be evaluated by simulating the effect of the proposed defense measures. Some or all of the above-mentioned processing in the defense unit may be performed, for example, using the generation AI, or may be performed without using the generation AI. For example, the defense unit can input data of the defense measures to the generation AI and have the generation AI execute a simulation of the effect.

[0037] The feedback unit can evaluate the effect of the defensive measures and generate a new attack scenario. It is necessary to clarify the specific content of the new attack scenario and the generation method. For example, the feedback unit can evaluate the effect of the defensive measures and generate a new attack scenario based on the result. For example, the feedback unit can generate an attack scenario that introduces a new attack method. The feedback unit can also generate an attack scenario that improves an existing scenario. In this way, a new attack scenario can be generated by evaluating the effect of the defensive measures. A part or all of the above-mentioned processing in the feedback unit may be performed using, for example, a generation AI, or may be performed without using the generation AI. For example, the feedback unit can input the effect data of the defensive measures to the generation AI and cause the generation AI to generate a new attack scenario.

[0038] The attacking unit can analyze the past attack history of the system and select the most effective vulnerability search method. It is necessary to clarify the specific content of the past attack history and the analysis method. For example, the attacking unit can analyze log data and search for similar vulnerabilities based on attack methods that have been successful in the past. The attacking unit can also prioritize searching for frequently targeted vulnerabilities from the past attack history. Furthermore, the attacking unit can analyze the past attack history and explore new attack methods. In this way, the most effective vulnerability search method can be selected by analyzing the past attack history. Part or all of the above-mentioned processing in the attacking unit may be performed, for example, using the generating AI, or may be performed without using the generating AI. For example, the attacking unit can input past attack history data to the generating AI and cause the generating AI to select a vulnerability search method.

[0039] The attacking unit can adjust the search method based on the operation status and load state of the system when searching for vulnerabilities. It is necessary to clarify the specific measurement method and criteria for the operation status and load state. For example, the attacking unit can monitor the CPU usage rate and memory usage, and use a lightweight search method when the system is in a high load state. In addition, the attacking unit can use a detailed search method when the system is in a low load state. Furthermore, the attacking unit can dynamically switch the search method according to the operation status of the system. This enables efficient vulnerability search by adjusting the search method based on the operation status and load state of the system. Part or all of the above-mentioned processing in the attacking unit may be performed, for example, using the generation AI, or may be performed without using the generation AI. For example, the attacking unit can input the operation status data of the system to the generation AI and cause the generation AI to adjust the search method.

[0040] The attacking unit can optimize the search method by considering the geographical distribution of the system when searching for vulnerabilities. It is necessary to clarify the specific content of the geographical distribution and the method of consideration. For example, the attacking unit can use a search method that considers the location of the data center and the location of the user and the characteristics of each region for a geographically distributed system. The attacking unit can also use a search method that simulates a concentrated attack for a geographically concentrated system. Furthermore, the attacking unit can dynamically adjust the search method based on the geographical distribution. In this way, the optimal search method can be selected by considering the geographical distribution of the system. Part or all of the above-mentioned processing in the attacking unit may be performed, for example, using the generating AI, or may be performed without using the generating AI. For example, the attacking unit can input the geographical distribution data of the system to the generating AI and cause the generating AI to execute optimization of the search method.

[0041] When searching for vulnerabilities, the attacking unit can improve the accuracy of the search by referring to related literature of the system. It is necessary to clarify the specific type of related literature and the method of reference. For example, the attacking unit can refer to the latest security research and update the vulnerability search method. The attacking unit can also refer to past security incident reports and optimize the vulnerability search method. Furthermore, the attacking unit can also strengthen the search method for a specific vulnerability by referring to related literature of the system. As a result, the accuracy of the search can be improved by referring to related literature of the system. Part or all of the above-mentioned processing in the attacking unit may be performed, for example, using the generating AI, or may be performed without using the generating AI. For example, the attacking unit can input related literature data to the generating AI and cause the generating AI to execute the improvement of the search accuracy.

[0042] When generating a defensive measure, the defense unit can select an optimal defensive measure by referring to past defense history. It is necessary to clarify the specific content of the past defense history and the method of referring to it. For example, the defense unit can analyze log data and select a defensive measure against a similar attack based on defensive measures that have been successful in the past. The defense unit can also preferentially select a frequently used defensive measure from the past defense history. Furthermore, the defense unit can analyze the past defense history and search for a new defensive measure. In this way, the optimal defensive measure can be selected by referring to the past defense history. A part or all of the above-mentioned processing in the defense unit may be performed, for example, using the generation AI, or may be performed without using the generation AI. For example, the defense unit can input past defense history data to the generation AI and cause the generation AI to select a defensive measure.

[0043] The defense unit can customize the defensive measures based on the current security policy of the system when generating the defensive measures. It is necessary to clarify the specific contents of the current security policy and the method of reference. For example, the defense unit can select an appropriate defensive measure by referring to the security guidelines and industry standards of the company. The defense unit can also dynamically adjust the defensive measures in response to changes in the security policy. Furthermore, the defense unit can also determine the priority of the defensive measures based on the security policy. As a result, it is possible to provide more effective defensive measures by customizing the defensive measures based on the current security policy. Some or all of the above-mentioned processing in the defense unit may be performed, for example, using the generation AI, or may be performed without using the generation AI. For example, the defense unit can input security policy data to the generation AI and cause the generation AI to execute customization of the defensive measures.

[0044] The defense unit can optimize the defense measures by considering the geographical distribution of the systems when generating the defense measures. It is necessary to clarify the specific content of the geographical distribution and the method of consideration. For example, the defense unit can consider the location of the data center and the location of the user, and propose defense measures that consider the characteristics of each region for a geographically distributed system. The defense unit can also propose defense measures to prevent concentrated attacks for a geographically concentrated system. Furthermore, the defense unit can dynamically adjust the defense measures based on the geographical distribution. In this way, it is possible to provide optimal defense measures by considering the geographical distribution of the systems. Part or all of the above-mentioned processing in the defense unit may be performed, for example, using the generation AI, or may be performed without using the generation AI. For example, the defense unit can input the geographical distribution data of the systems to the generation AI and cause the generation AI to execute optimization of the defense measures.

[0045] When generating a defensive measure, the defense unit can improve the accuracy of the defensive measure by referring to related literature of the system. It is necessary to clarify the specific type of related literature and the method of reference. For example, the defense unit can refer to the latest security research and update the defensive measure. In addition, the defense unit can also refer to past security incident reports and optimize the defensive measure. Furthermore, the defense unit can also strengthen the defensive measure against a specific attack by referring to related literature of the system. As a result, the accuracy of the defensive measure can be improved by referring to related literature of the system. Part or all of the above-mentioned processing in the defense unit may be performed, for example, using the generation AI, or may be performed without using the generation AI. For example, the defense unit can input related literature data to the generation AI and cause the generation AI to execute the improvement of the accuracy of the defensive measure.

[0046] At the time of feedback, the feedback unit can select the optimal feedback method by referring to the past feedback history. It is necessary to clarify the specific content of the past feedback history and the method of referring to it. For example, the feedback unit can analyze the log data and select feedback for a similar situation based on a feedback method that was successful in the past. The feedback unit can also preferentially select a frequently used feedback method from the past feedback history. Furthermore, the feedback unit can analyze the past feedback history and search for a new feedback method. In this way, the optimal feedback method can be selected by referring to the past feedback history. A part or all of the above-mentioned processing in the feedback unit may be performed, for example, using the generation AI, or may be performed without using the generation AI. For example, the feedback unit can input the past feedback history data to the generation AI and cause the generation AI to select the feedback method.

[0047] The feedback unit can customize the feedback content based on the current state of the system at the time of feedback. It is necessary to clarify the specific content of the current state and the measurement method. For example, the feedback unit can monitor the operation status and security level of the system and provide appropriate feedback based on the current system state. The feedback unit can also dynamically adjust the feedback content according to changes in the system state. Furthermore, the feedback unit can also determine the priority of the feedback based on the system state. As a result, more appropriate feedback can be provided by customizing the feedback content based on the current state of the system. A part or all of the above-mentioned processing in the feedback unit may be performed, for example, using the generation AI, or may be performed without using the generation AI. For example, the feedback unit can input system state data to the generation AI and cause the generation AI to customize the feedback content.

[0048] The feedback unit can optimize the feedback method by taking into account the geographical distribution of the system at the time of feedback. It is necessary to clarify the specific content of the geographical distribution and the method of consideration. For example, the feedback unit can provide feedback that takes into account the location of the data center and the location of the user, to a geographically distributed system, taking into account the characteristics of each region. The feedback unit can also provide feedback that prevents concentrated attacks to a geographically concentrated system. Furthermore, the feedback unit can dynamically adjust the feedback method based on the geographical distribution. In this way, an optimal feedback method can be provided by taking into account the geographical distribution of the system. A part or all of the above-mentioned processing in the feedback unit may be performed, for example, using a generation AI, or may be performed without using a generation AI. For example, the feedback unit can input the geographical distribution data of the system to the generation AI and cause the generation AI to execute optimization of the feedback method.

[0049] The feedback unit can improve the accuracy of the feedback by referring to the related literature of the system at the time of feedback. It is necessary to clarify the specific type of related literature and the reference method. For example, the feedback unit can refer to the latest security research and update the feedback content. The feedback unit can also refer to past security incident reports and optimize the feedback content. Furthermore, the feedback unit can also strengthen the feedback for a specific attack by referring to the related literature of the system. As a result, the accuracy of the feedback can be improved by referring to the related literature of the system. A part or all of the above-mentioned processing in the feedback unit may be performed, for example, using the generation AI, or may be performed without using the generation AI. For example, the feedback unit can input related literature data to the generation AI and cause the generation AI to execute the improvement of the accuracy of the feedback.

[0050] The system according to the embodiment is not limited to the above-described example, and various modifications are possible, for example, as follows.

[0051] When searching for system vulnerabilities, the attacking unit can analyze user operation logs in real time and prioritize the discovery of vulnerabilities based on the user's behavioral patterns. For example, it is possible to prioritize the search for vulnerabilities in functions or pages that are frequently accessed by users. Also, if it is found from the user operation logs that a particular operation occurs frequently, it is possible to prioritize the search for vulnerabilities related to that operation. Furthermore, by analyzing the user operation logs, it is also possible to prioritize the discovery of vulnerabilities during a particular time period. This makes it possible to efficiently discover vulnerabilities based on the user's behavioral patterns.

[0052] When searching for system vulnerabilities, the attacking unit can automatically collect system configuration information and prioritize the discovery of vulnerabilities based on the configuration information. For example, it is possible to identify from the system configuration information that a particular software version or setting is vulnerable, and to prioritize the discovery of that part. In addition, by analyzing the system configuration information, it is also possible to prioritize the discovery of vulnerabilities in a particular network segment or server. Furthermore, it is also possible to prioritize the discovery of vulnerabilities in a particular device or application based on the system configuration information. This makes it possible to efficiently discover vulnerabilities based on the system configuration information.

[0053] When searching for system vulnerabilities, the attacking department can refer to external threat intelligence data and prioritize the discovery of vulnerabilities based on the latest threat information. For example, the attacking department can obtain the latest attack techniques and vulnerability information from external threat intelligence data and search for vulnerabilities based on that information. In addition, by analyzing external threat intelligence data, the attacking department can prioritize the discovery of vulnerabilities against techniques used by a specific attacker group. Furthermore, the attacking department can prioritize the discovery of vulnerabilities against a specific industry or region based on external threat intelligence data. This makes it possible to efficiently discover vulnerabilities based on the latest threat information.

[0054] When analyzing an attack scenario, the defense unit can refer to the system's past defense history and give priority to proposing defensive measures that were effective in the past. For example, it is possible to identify defensive measures that were effective against a specific attack method from the past defense history and give priority to proposing those defensive measures. In addition, by analyzing the past defense history, it is also possible to give priority to proposing defensive measures that were effective for a specific system configuration or setting. Furthermore, it is also possible to give priority to proposing defensive measures that were effective for a specific time period or situation based on the past defense history. This makes it possible to efficiently propose defensive measures based on the past defense history.

[0055] When simulating the effect of a proposed defensive measure, the defense unit can perform the simulation taking into consideration the current operating status and load state of the system. For example, when the system is in a high-load state, it is possible to give priority to simulating a lightweight defensive measure. Also, when the system is in a low-load state, it is possible to simulate a detailed defensive measure. Furthermore, it is possible to dynamically adjust the priority of the simulation according to the operating status and load state of the system. This makes it possible to efficiently simulate the defensive measures based on the operating status and load state of the system.

[0056] The process flow of the first embodiment will be briefly described below.

[0057] Step 1: The attack unit searches for vulnerabilities in the system. The attack unit performs a detailed analysis of each component of the system to identify potential vulnerabilities. For example, the attack unit can identify unupdated software and improper settings as vulnerabilities. The attack unit also generates attack scenarios based on the identified vulnerabilities. For example, the attack unit can generate attack scenarios that include attack methods such as SQL injection and cross-site scripting (XSS). Step 2: The defense unit generates defense measures based on the attack scenario. The defense unit analyzes the attack scenario and proposes optimal defense measures. For example, the defense unit can propose changes to firewall settings or software updates as defense measures. The defense unit also simulates the effectiveness of the proposed defense measures and evaluates their actual effectiveness. For example, the defense unit can test the effectiveness of the defense measures in a virtual environment and evaluate the results. Step 3: The feedback unit evaluates the effectiveness of the defensive measures and generates new attack scenarios. The feedback unit can evaluate the effectiveness of the defensive measures and generate new attack scenarios based on the results. This enables the penetration testing support system to strengthen the security of the system from both the offensive and defensive perspectives.

[0058] (Example 2) A penetration test support system according to an embodiment of the present invention is a system that supports penetration tests using a generation AI that has been trained in two patterns, one for an attack team and one for a defense team. In this system, the generation AI for the attack team searches for vulnerabilities in the system and generates an attack scenario. Next, the generation AI for the defense team generates a defense measure against the attack scenario. This makes it possible to strengthen the security of the system from both the attack and defense perspectives. For example, the generation AI for the attack team analyzes each component of the system in detail and identifies potential vulnerabilities. For example, unupdated software and improper settings are examples of vulnerabilities. Next, the generation AI generates an attack scenario based on the identified vulnerability. For example, attack methods such as SQL injection and cross-site scripting (XSS) are included. Next, the generation AI for the defense team generates a defense measure against the attack scenario. The generation AI analyzes the attack scenario and proposes an optimal defense measure. For example, firewall setting changes and software updates are examples of defense measures. Furthermore, the generation AI for the defense team simulates the effect of the proposed defense measure and evaluates the actual effect. Finally, the generation AI for the attack team and the generation AI for the defense team provide feedback to each other. The generation AI for the attack team evaluates the effectiveness of defensive measures and generates new attack scenarios. Meanwhile, the generation AI for the defense team again generates defensive measures against new attack scenarios. By repeating this process, the security of the system can be continuously improved. This makes it possible to strengthen the security of the system from both sides of attack and defense. For example, if the generation AI for the attack team discovers a new vulnerability, the generation AI for the defense team can immediately propose countermeasures, quickly improving the security of the system. In addition, through a feedback loop, attack and defense skills are improved, enabling more advanced security measures. This allows the penetration test support system to strengthen the security of the system from both sides of attack and defense.

[0059] A penetration test support system according to an embodiment includes an attack unit, a defense unit, and a feedback unit. The attack unit searches for vulnerabilities in the system. For example, the attack unit analyzes each component of the system in detail to identify potential vulnerabilities. For example, the attack unit can identify unupdated software, improper settings, and the like as vulnerabilities. The attack unit also generates an attack scenario based on the identified vulnerabilities. For example, the attack unit can generate an attack scenario including an attack method such as SQL injection or cross-site scripting (XSS). The defense unit generates a defense measure based on the attack scenario. The defense unit, for example, analyzes the attack scenario and proposes an optimal defense measure. For example, the defense unit can propose a firewall setting change, software update, and the like as a defense measure. The defense unit also simulates the effect of the proposed defense measure and evaluates the actual effect. For example, the defense unit can test the effect of the defense measure in a virtual environment and evaluate the result. The feedback unit evaluates the effect of the defense measure and generates a new attack scenario. The feedback unit can evaluate the effect of the defense measure and generate a new attack scenario based on the result. As a result, the penetration test support system according to the embodiment can strengthen the security of the system from both the attack and defense perspectives. For example, the attack unit searches for vulnerabilities in the system and generates attack scenarios. The defense unit generates defensive measures based on the attack scenarios and evaluates their effectiveness. The feedback unit evaluates the effectiveness of the defensive measures and generates new attack scenarios. As a result, the penetration test support system can continuously improve the security of the system.

[0060] The attack unit searches for vulnerabilities in the system. For example, the attack unit analyzes each component of the system in detail to identify potential vulnerabilities. Specifically, the attack unit analyzes the source code of the system using a static analysis tool to detect vulnerabilities in the code. In addition, the attack unit can use a dynamic analysis tool to monitor the behavior of the running system and detect abnormal behavior and unauthorized access. Furthermore, the attack unit uses a network scanner to analyze the network configuration of the system and identify open ports and services. This makes it possible to identify unupdated software and improper settings as vulnerabilities. In addition, the attack unit generates attack scenarios based on the identified vulnerabilities. For example, the attack unit can generate attack scenarios that include attack methods such as SQL injection and cross-site scripting (XSS). These attack scenarios are used to simulate actual attacks and are an important means for evaluating the defensive power of the system. The attack unit executes the generated attack scenarios and evaluates the impact by actually attacking the vulnerabilities of the system. This allows the attack unit to grasp the vulnerabilities of the system in detail and verify the vulnerabilities using specific attack methods.

[0061] The defense unit generates a defense measure based on the attack scenario. For example, the defense unit analyzes the attack scenario and proposes an optimal defense measure. Specifically, the defense unit analyzes the contents of the attack scenario in detail and identifies the attack method and attack path. This makes it possible to clarify the source of the attack and the purpose of the attack, and to formulate optimal defense measures against them. For example, the defense unit can propose a firewall setting change or software update as a defense measure. By changing the firewall setting, it is possible to restrict access from a specific port or IP address and prevent unauthorized access. In addition, by updating the software, it is possible to fix known vulnerabilities and strengthen the security of the system. Furthermore, the defense unit simulates the effect of the proposed defense measure and evaluates the actual effect. For example, the defense unit can test the effect of the defense measure in a virtual environment and evaluate the result. In the virtual environment, it is possible to verify the effect of the defense measure without affecting the actual system. This allows the defense unit to check whether the proposed defense measure is actually effective and to modify the defense measure as necessary. The defense unit can continuously review the defense measures and provide optimal defense measures in order to strengthen the security of the system.

[0062] The feedback unit evaluates the effectiveness of the defensive measures and generates a new attack scenario. For example, the feedback unit can evaluate the effectiveness of the defensive measures and generate a new attack scenario based on the evaluation result. Specifically, the feedback unit monitors the security state of the system after the defensive measures are implemented and evaluates how effective the defensive measures are. For example, the feedback unit analyzes the results of an attack simulation performed after the defensive measures are implemented and evaluates how much the defensive measures have prevented attacks. This makes it possible to confirm the effectiveness of the defensive measures and to modify the defensive measures as necessary. Furthermore, the feedback unit generates a new attack scenario based on the evaluation result of the effectiveness of the defensive measures. For example, if the defensive measures are effective against a specific attack method, a new attack scenario that avoids the attack method can be generated. This makes it possible to continuously strengthen the security of the system. Furthermore, the feedback unit can accumulate past attack scenarios and the effects of defensive measures in a database and utilize them for generating future attack scenarios and formulating defensive measures. This makes it possible for the feedback unit to continuously improve the security of the system and strengthen the security of the system from both attack and defense perspectives.

[0063] The attacking part can analyze each component of the system in detail to identify potential vulnerabilities. It is necessary to clarify the specific method and criteria for the detailed analysis. For example, the attacking part can perform a code review to identify vulnerabilities in the source code. The attacking part can also perform a penetration test to identify vulnerabilities in the system. For example, the attacking part can analyze each component of the system in detail to identify unfixed bugs and known security holes. As a result, potential vulnerabilities can be identified by analyzing each component of the system in detail. A part or all of the above-mentioned processing in the attacking part may be performed using, for example, a generation AI, or may be performed without using a generation AI. For example, the attacking part can input data of each component of the system into the generation AI and have the generation AI execute the identification of vulnerabilities.

[0064] The attack unit can generate an attack scenario based on the identified vulnerability. It is necessary to clarify the specific type of the identified vulnerability and the method of identification. For example, the attack unit can identify vulnerabilities such as SQL injection and cross-site scripting (XSS) and generate an attack scenario based on the vulnerability. It is necessary to clarify the specific content of the attack scenario and the method of generation. For example, the attack unit can generate an SQL injection attack scenario based on the identified vulnerability. The attack unit can also generate a cross-site scripting (XSS) attack scenario. As a result, an effective attack scenario can be created by generating an attack scenario based on the identified vulnerability. A part or all of the above-mentioned processing in the attack unit may be performed using, for example, a generation AI, or may be performed without using the generation AI. For example, the attack unit can input data of the identified vulnerability to the generation AI and cause the generation AI to generate an attack scenario.

[0065] The defense unit can analyze the attack scenario and propose appropriate defense measures. It is necessary to clarify the specific content of the appropriate defense measures and the method of proposing them. For example, the defense unit can analyze the attack scenario and propose a change in firewall settings. The defense unit can also propose software updates. In this way, the optimal defense measures can be proposed by analyzing the attack scenario. Some or all of the above-mentioned processing in the defense unit may be performed, for example, using the generation AI, or may be performed without using the generation AI. For example, the defense unit can input data of the attack scenario to the generation AI and have the generation AI execute the proposal of the defense measures.

[0066] The defense unit can simulate the effect of the proposed defense measures and evaluate the actual effect. It is necessary to clarify the specific method and criteria for simulating the effect. For example, the defense unit can test the effect of the defense measures in a virtual environment and evaluate the result. The defense unit can also perform a scenario-based simulation and evaluate the effect of the defense measures. In this way, the actual effect can be evaluated by simulating the effect of the proposed defense measures. Some or all of the above-mentioned processing in the defense unit may be performed, for example, using the generation AI, or may be performed without using the generation AI. For example, the defense unit can input data of the defense measures to the generation AI and have the generation AI execute a simulation of the effect.

[0067] The feedback unit can evaluate the effect of the defensive measures and generate a new attack scenario. It is necessary to clarify the specific content of the new attack scenario and the generation method. For example, the feedback unit can evaluate the effect of the defensive measures and generate a new attack scenario based on the result. For example, the feedback unit can generate an attack scenario that introduces a new attack method. The feedback unit can also generate an attack scenario that improves an existing scenario. In this way, a new attack scenario can be generated by evaluating the effect of the defensive measures. A part or all of the above-mentioned processing in the feedback unit may be performed using, for example, a generation AI, or may be performed without using the generation AI. For example, the feedback unit can input the effect data of the defensive measures to the generation AI and cause the generation AI to generate a new attack scenario.

[0068] The attacking unit can estimate the user's emotions and adjust the priority of vulnerability search based on the estimated user's emotions. It is necessary to clarify how to estimate the user's emotions. For example, the attacking unit can conduct a questionnaire survey to estimate the user's emotions. The attacking unit can also perform behavioral analysis to estimate the user's emotions. For example, the attacking unit can prioritize searching for the most serious vulnerabilities when the user feels anxious. The attacking unit can also search for overall vulnerabilities in a balanced manner when the user is relaxed. Furthermore, the attacking unit can prioritize searching for vulnerabilities that can be found quickly when the user is in a hurry. This enables more effective vulnerability search by adjusting the priority of vulnerability search based on the user's emotions. The estimation of emotions is realized using an emotion estimation function using, for example, an emotion engine or a generation AI. The generation AI is, for example, a text generation AI (for example, LLM) or a multimodal generation AI, but is not limited to such examples. A part or all of the above-mentioned processing in the attacking unit may be performed using, for example, a generation AI, or may be performed without using a generation AI. For example, the attacking unit can input user emotional data into the generation AI and have the generation AI adjust the priority of vulnerability searches.

[0069] The attacking unit can analyze the past attack history of the system and select the most effective vulnerability search method. It is necessary to clarify the specific content of the past attack history and the analysis method. For example, the attacking unit can analyze log data and search for similar vulnerabilities based on attack methods that have been successful in the past. The attacking unit can also prioritize searching for frequently targeted vulnerabilities from the past attack history. Furthermore, the attacking unit can analyze the past attack history and explore new attack methods. In this way, the most effective vulnerability search method can be selected by analyzing the past attack history. Part or all of the above-mentioned processing in the attacking unit may be performed, for example, using the generating AI, or may be performed without using the generating AI. For example, the attacking unit can input past attack history data to the generating AI and cause the generating AI to select a vulnerability search method.

[0070] The attacking unit can adjust the search method based on the operation status and load state of the system when searching for vulnerabilities. It is necessary to clarify the specific measurement method and criteria for the operation status and load state. For example, the attacking unit can monitor the CPU usage rate and memory usage, and use a lightweight search method when the system is in a high load state. In addition, the attacking unit can use a detailed search method when the system is in a low load state. Furthermore, the attacking unit can dynamically switch the search method according to the operation status of the system. This enables efficient vulnerability search by adjusting the search method based on the operation status and load state of the system. Part or all of the above-mentioned processing in the attacking unit may be performed, for example, using the generation AI, or may be performed without using the generation AI. For example, the attacking unit can input the operation status data of the system to the generation AI and cause the generation AI to adjust the search method.

[0071] The attacking unit can estimate the user's emotions and adjust the vulnerability reporting method based on the estimated user's emotions. It is necessary to clarify how to estimate the user's emotions. For example, the attacking unit can conduct a questionnaire survey to estimate the user's emotions. The attacking unit can also perform behavioral analysis to estimate the user's emotions. For example, the attacking unit can provide a detailed report to provide a sense of security when the user feels anxious. The attacking unit can also provide a concise report to encourage a quick response when the user is relaxed. Furthermore, the attacking unit can also provide a report that focuses on the main points to encourage a quick response when the user is in a hurry. This allows for more appropriate reporting by adjusting the vulnerability reporting method based on the user's emotions. The emotion estimation is realized using an emotion estimation function using, for example, an emotion engine or a generation AI. The generation AI is, for example, a text generation AI (for example, LLM) or a multimodal generation AI, but is not limited to such examples. A part or all of the above-mentioned processing in the attacking unit may be performed using, for example, a generation AI, or may be performed without using a generation AI. For example, the attacking department can input user emotional data into the generation AI and have the generation AI adjust the way vulnerabilities are reported.

[0072] The attacking unit can optimize the search method by considering the geographical distribution of the system when searching for vulnerabilities. It is necessary to clarify the specific content of the geographical distribution and the method of consideration. For example, the attacking unit can use a search method that considers the location of the data center and the location of the user and the characteristics of each region for a geographically distributed system. The attacking unit can also use a search method that simulates a concentrated attack for a geographically concentrated system. Furthermore, the attacking unit can dynamically adjust the search method based on the geographical distribution. In this way, the optimal search method can be selected by considering the geographical distribution of the system. Part or all of the above-mentioned processing in the attacking unit may be performed, for example, using the generating AI, or may be performed without using the generating AI. For example, the attacking unit can input the geographical distribution data of the system to the generating AI and cause the generating AI to execute optimization of the search method.

[0073] When searching for vulnerabilities, the attacking unit can improve the accuracy of the search by referring to related literature of the system. It is necessary to clarify the specific type of related literature and the method of reference. For example, the attacking unit can refer to the latest security research and update the vulnerability search method. The attacking unit can also refer to past security incident reports and optimize the vulnerability search method. Furthermore, the attacking unit can also strengthen the search method for a specific vulnerability by referring to related literature of the system. As a result, the accuracy of the search can be improved by referring to related literature of the system. Part or all of the above-mentioned processing in the attacking unit may be performed, for example, using the generating AI, or may be performed without using the generating AI. For example, the attacking unit can input related literature data to the generating AI and cause the generating AI to execute the improvement of the search accuracy.

[0074] The defense unit can estimate the user's emotions and adjust the method of proposing defensive measures based on the estimated user's emotions. It is necessary to clarify how to estimate the user's emotions. For example, the defense unit can conduct a questionnaire survey and estimate the user's emotions. The defense unit can also perform behavioral analysis and estimate the user's emotions. For example, when the user feels anxious, the defense unit can propose detailed defensive measures to provide a sense of security. When the user is relaxed, the defense unit can also propose concise defensive measures and encourage a quick response. Furthermore, when the user is in a hurry, the defense unit can also propose a defensive measure that focuses on the main points and encourage a quick response. This makes it possible to propose more appropriate defensive measures by adjusting the method of proposing defensive measures based on the user's emotions. The estimation of emotions is realized using an emotion estimation function using, for example, an emotion engine or a generation AI. The generation AI is, for example, a text generation AI (for example, LLM) or a multimodal generation AI, but is not limited to such examples. A part or all of the above-mentioned processing in the defense unit may be performed using, for example, a generation AI, or may be performed without using a generation AI. For example, the defense unit can input a user's emotional data into the generation AI and have the generation AI adjust the method of proposing defense measures.

[0075] When generating a defensive measure, the defense unit can select an optimal defensive measure by referring to past defense history. It is necessary to clarify the specific content of the past defense history and the method of referring to it. For example, the defense unit can analyze log data and select a defensive measure against a similar attack based on defensive measures that have been successful in the past. The defense unit can also preferentially select a frequently used defensive measure from the past defense history. Furthermore, the defense unit can analyze the past defense history and search for a new defensive measure. In this way, the optimal defensive measure can be selected by referring to the past defense history. A part or all of the above-mentioned processing in the defense unit may be performed, for example, using the generation AI, or may be performed without using the generation AI. For example, the defense unit can input past defense history data to the generation AI and cause the generation AI to select a defensive measure.

[0076] The defense unit can customize the defensive measures based on the current security policy of the system when generating the defensive measures. It is necessary to clarify the specific contents of the current security policy and the method of reference. For example, the defense unit can select an appropriate defensive measure by referring to the security guidelines and industry standards of the company. The defense unit can also dynamically adjust the defensive measures in response to changes in the security policy. Furthermore, the defense unit can also determine the priority of the defensive measures based on the security policy. As a result, it is possible to provide more effective defensive measures by customizing the defensive measures based on the current security policy. Some or all of the above-mentioned processing in the defense unit may be performed, for example, using the generation AI, or may be performed without using the generation AI. For example, the defense unit can input security policy data to the generation AI and cause the generation AI to execute customization of the defensive measures.

[0077] The defense unit can estimate the user's emotions and determine the priority of defense measures based on the estimated user's emotions. It is necessary to clarify how to estimate the user's emotions. For example, the defense unit can conduct a questionnaire survey and estimate the user's emotions. The defense unit can also perform behavioral analysis to estimate the user's emotions. For example, the defense unit can prioritize defense measures against the most serious attacks when the user feels anxious. The defense unit can also propose overall defense measures in a balanced manner when the user is relaxed. Furthermore, the defense unit can also prioritize defense measures that can be quickly implemented when the user is in a hurry. In this way, by determining the priority of defense measures based on the user's emotions, more effective defense measures can be provided. The estimation of emotions is realized using an emotion estimation function using, for example, an emotion engine or a generation AI. The generation AI is, for example, a text generation AI (for example, LLM) or a multimodal generation AI, but is not limited to such examples. A part or all of the above-mentioned processing in the defense unit may be performed using, for example, a generation AI, or may be performed without using a generation AI. For example, the defense unit can input a user's emotional data into the generation AI and have the generation AI determine the priority of defense measures.

[0078] The defense unit can optimize the defense measures by considering the geographical distribution of the systems when generating the defense measures. It is necessary to clarify the specific content of the geographical distribution and the method of consideration. For example, the defense unit can consider the location of the data center and the location of the user, and propose defense measures that consider the characteristics of each region for a geographically distributed system. The defense unit can also propose defense measures to prevent concentrated attacks for a geographically concentrated system. Furthermore, the defense unit can dynamically adjust the defense measures based on the geographical distribution. In this way, it is possible to provide optimal defense measures by considering the geographical distribution of the systems. Part or all of the above-mentioned processing in the defense unit may be performed, for example, using the generation AI, or may be performed without using the generation AI. For example, the defense unit can input the geographical distribution data of the systems to the generation AI and cause the generation AI to execute optimization of the defense measures.

[0079] When generating a defensive measure, the defense unit can improve the accuracy of the defensive measure by referring to related literature of the system. It is necessary to clarify the specific type of related literature and the method of reference. For example, the defense unit can refer to the latest security research and update the defensive measure. In addition, the defense unit can also refer to past security incident reports and optimize the defensive measure. Furthermore, the defense unit can also strengthen the defensive measure against a specific attack by referring to related literature of the system. As a result, the accuracy of the defensive measure can be improved by referring to related literature of the system. Part or all of the above-mentioned processing in the defense unit may be performed, for example, using the generation AI, or may be performed without using the generation AI. For example, the defense unit can input related literature data to the generation AI and cause the generation AI to execute the improvement of the accuracy of the defensive measure.

[0080] The feedback unit can estimate the user's emotion and adjust the feedback method based on the estimated user's emotion. It is necessary to clarify how to estimate the user's emotion. For example, the feedback unit can conduct a questionnaire survey and estimate the user's emotion. The feedback unit can also perform behavioral analysis and estimate the user's emotion. For example, the feedback unit can provide detailed feedback and provide a sense of security when the user feels anxious. The feedback unit can also provide concise feedback and encourage a quick response when the user is relaxed. Furthermore, the feedback unit can provide feedback that focuses on the main points and encourage a quick response when the user is in a hurry. This makes it possible to provide more appropriate feedback by adjusting the feedback method based on the user's emotion. The emotion estimation is realized using an emotion estimation function using, for example, an emotion engine or a generation AI. The generation AI is, for example, a text generation AI (for example, LLM) or a multimodal generation AI, but is not limited to such examples. A part or all of the above-mentioned processing in the feedback unit may be performed using, for example, a generation AI, or may be performed without using a generation AI. For example, the feedback unit can input the user's emotional data into the generation AI and cause the generation AI to adjust the feedback method.

[0081] At the time of feedback, the feedback unit can select the optimal feedback method by referring to the past feedback history. It is necessary to clarify the specific content of the past feedback history and the method of referring to it. For example, the feedback unit can analyze the log data and select feedback for a similar situation based on a feedback method that was successful in the past. The feedback unit can also preferentially select a frequently used feedback method from the past feedback history. Furthermore, the feedback unit can analyze the past feedback history and search for a new feedback method. In this way, the optimal feedback method can be selected by referring to the past feedback history. A part or all of the above-mentioned processing in the feedback unit may be performed, for example, using the generation AI, or may be performed without using the generation AI. For example, the feedback unit can input the past feedback history data to the generation AI and cause the generation AI to select the feedback method.

[0082] The feedback unit can customize the feedback content based on the current state of the system at the time of feedback. It is necessary to clarify the specific content of the current state and the measurement method. For example, the feedback unit can monitor the operation status and security level of the system and provide appropriate feedback based on the current system state. The feedback unit can also dynamically adjust the feedback content according to changes in the system state. Furthermore, the feedback unit can also determine the priority of the feedback based on the system state. As a result, more appropriate feedback can be provided by customizing the feedback content based on the current state of the system. A part or all of the above-mentioned processing in the feedback unit may be performed, for example, using the generation AI, or may be performed without using the generation AI. For example, the feedback unit can input system state data to the generation AI and cause the generation AI to customize the feedback content.

[0083] The feedback unit can estimate the user's emotion and determine the priority of feedback based on the estimated user's emotion. It is necessary to clarify how to estimate the user's emotion. For example, the feedback unit can conduct a questionnaire survey and estimate the user's emotion. The feedback unit can also perform behavioral analysis and estimate the user's emotion. For example, the feedback unit can prioritize the most important feedback when the user feels anxious. The feedback unit can also provide overall feedback in a balanced manner when the user is relaxed. Furthermore, the feedback unit can also prioritize feedback that can be quickly responded to when the user is in a hurry. In this way, by determining the priority of feedback based on the user's emotion, more effective feedback can be provided. The emotion estimation is realized using an emotion estimation function using, for example, an emotion engine or a generation AI. The generation AI is, for example, a text generation AI (for example, LLM) or a multimodal generation AI, but is not limited to such examples. A part or all of the above-mentioned processing in the feedback unit may be performed using, for example, a generation AI, or may be performed without using a generation AI. For example, the feedback unit can input the user's emotional data into the generation AI and have the generation AI determine the priority of the feedback.

[0084] The feedback unit can optimize the feedback method by taking into account the geographical distribution of the system at the time of feedback. It is necessary to clarify the specific content of the geographical distribution and the method of consideration. For example, the feedback unit can provide feedback that takes into account the location of the data center and the location of the user, to a geographically distributed system, taking into account the characteristics of each region. The feedback unit can also provide feedback that prevents concentrated attacks to a geographically concentrated system. Furthermore, the feedback unit can dynamically adjust the feedback method based on the geographical distribution. In this way, an optimal feedback method can be provided by taking into account the geographical distribution of the system. A part or all of the above-mentioned processing in the feedback unit may be performed, for example, using a generation AI, or may be performed without using a generation AI. For example, the feedback unit can input the geographical distribution data of the system to the generation AI and cause the generation AI to execute optimization of the feedback method.

[0085] The feedback unit can improve the accuracy of the feedback by referring to the related literature of the system at the time of feedback. It is necessary to clarify the specific type of related literature and the reference method. For example, the feedback unit can refer to the latest security research and update the feedback content. The feedback unit can also refer to past security incident reports and optimize the feedback content. Furthermore, the feedback unit can also strengthen the feedback for a specific attack by referring to the related literature of the system. As a result, the accuracy of the feedback can be improved by referring to the related literature of the system. A part or all of the above-mentioned processing in the feedback unit may be performed, for example, using the generation AI, or may be performed without using the generation AI. For example, the feedback unit can input related literature data to the generation AI and cause the generation AI to execute the improvement of the accuracy of the feedback.

[0086] The system according to the embodiment is not limited to the above-described example, and various modifications are possible, for example, as follows.

[0087] When searching for system vulnerabilities, the attacking unit can analyze user operation logs in real time and prioritize the discovery of vulnerabilities based on the user's behavioral patterns. For example, it is possible to prioritize the search for vulnerabilities in functions or pages that are frequently accessed by users. Also, if it is found from the user operation logs that a particular operation occurs frequently, it is possible to prioritize the search for vulnerabilities related to that operation. Furthermore, by analyzing the user operation logs, it is also possible to prioritize the discovery of vulnerabilities during a particular time period. This makes it possible to efficiently discover vulnerabilities based on the user's behavioral patterns.

[0088] When searching for system vulnerabilities, the attacking unit can automatically collect system configuration information and prioritize the discovery of vulnerabilities based on the configuration information. For example, it is possible to identify from the system configuration information that a particular software version or setting is vulnerable, and to prioritize the discovery of that part. In addition, by analyzing the system configuration information, it is also possible to prioritize the discovery of vulnerabilities in a particular network segment or server. Furthermore, it is also possible to prioritize the discovery of vulnerabilities in a particular device or application based on the system configuration information. This makes it possible to efficiently discover vulnerabilities based on the system configuration information.

[0089] When searching for system vulnerabilities, the attacking department can refer to external threat intelligence data and prioritize the discovery of vulnerabilities based on the latest threat information. For example, the attacking department can obtain the latest attack techniques and vulnerability information from external threat intelligence data and search for vulnerabilities based on that information. In addition, by analyzing external threat intelligence data, the attacking department can prioritize the discovery of vulnerabilities against techniques used by a specific attacker group. Furthermore, the attacking department can prioritize the discovery of vulnerabilities against a specific industry or region based on external threat intelligence data. This makes it possible to efficiently discover vulnerabilities based on the latest threat information.

[0090] When analyzing an attack scenario, the defense unit can refer to the system's past defense history and give priority to proposing defensive measures that were effective in the past. For example, it is possible to identify defensive measures that were effective against a specific attack method from the past defense history and give priority to proposing those defensive measures. In addition, by analyzing the past defense history, it is also possible to give priority to proposing defensive measures that were effective for a specific system configuration or setting. Furthermore, it is also possible to give priority to proposing defensive measures that were effective for a specific time period or situation based on the past defense history. This makes it possible to efficiently propose defensive measures based on the past defense history.

[0091] When simulating the effect of a proposed defensive measure, the defense unit can perform the simulation taking into consideration the current operating status and load state of the system. For example, when the system is in a high-load state, it is possible to give priority to simulating a lightweight defensive measure. Also, when the system is in a low-load state, it is possible to simulate a detailed defensive measure. Furthermore, it is possible to dynamically adjust the priority of the simulation according to the operating status and load state of the system. This makes it possible to efficiently simulate the defensive measures based on the operating status and load state of the system.

[0092] When evaluating the effectiveness of the defensive measures, the feedback unit can estimate the user's emotions and adjust the evaluation criteria based on the estimated user's emotions. For example, when the user feels anxious, the effectiveness of the defensive measures can be evaluated using detailed evaluation criteria. When the user feels relaxed, the effectiveness of the defensive measures can be evaluated using simple evaluation criteria. Furthermore, when the user is in a hurry, the effectiveness of the defensive measures can be evaluated using criteria that allow for quick evaluation. This makes it possible to efficiently evaluate the effectiveness of the defensive measures based on the user's emotions.

[0093] The attack unit can estimate the user's emotions and adjust the priority of vulnerability search based on the estimated user's emotions. For example, if the user feels anxious, the most serious vulnerabilities can be searched for preferentially. Also, if the user feels relaxed, overall vulnerabilities can be searched for in a balanced manner. Furthermore, if the user is in a hurry, vulnerabilities that can be found quickly can be searched for preferentially. In this way, by adjusting the priority of vulnerability search based on the user's emotions, more effective vulnerability search is possible.

[0094] The defense unit can estimate the user's emotions and adjust the method of proposing defensive measures based on the estimated user's emotions. For example, if the user feels anxious, detailed defensive measures can be proposed to provide a sense of security. If the user feels relaxed, concise defensive measures can be proposed to encourage a quick response. If the user is in a hurry, a defensive measure that focuses on the main points can be proposed to encourage a quick response. In this way, by adjusting the method of proposing defensive measures based on the user's emotions, more appropriate defensive measures can be proposed.

[0095] The feedback unit can estimate the user's emotion and adjust the feedback method based on the estimated user's emotion. For example, if the user feels anxious, detailed feedback can be provided to provide a sense of security. If the user is relaxed, brief feedback can be provided to encourage a quick response. If the user is in a hurry, feedback that focuses on the main points can be provided to encourage a quick response. In this way, more appropriate feedback can be provided by adjusting the feedback method based on the user's emotion.

[0096] The defense unit can estimate the user's emotions and prioritize defense measures based on the estimated user's emotions. For example, if the user feels anxious, the defense measures against the most serious attacks can be prioritized. Also, if the user feels relaxed, overall defense measures can be proposed in a balanced manner. Furthermore, if the user is in a hurry, defense measures that can be quickly implemented can be prioritized. In this way, by prioritizing defense measures based on the user's emotions, more effective defense measures can be provided.

[0097] The process flow of the second embodiment will be briefly described below.

[0098] Step 1: The attack unit searches for vulnerabilities in the system. The attack unit performs a detailed analysis of each component of the system to identify potential vulnerabilities. For example, the attack unit can identify unupdated software and improper settings as vulnerabilities. The attack unit also generates attack scenarios based on the identified vulnerabilities. For example, the attack unit can generate attack scenarios that include attack methods such as SQL injection and cross-site scripting (XSS). Step 2: The defense unit generates defense measures based on the attack scenario. The defense unit analyzes the attack scenario and proposes optimal defense measures. For example, the defense unit can propose changes to firewall settings or software updates as defense measures. The defense unit also simulates the effectiveness of the proposed defense measures and evaluates their actual effectiveness. For example, the defense unit can test the effectiveness of the defense measures in a virtual environment and evaluate the results. Step 3: The feedback unit evaluates the effectiveness of the defensive measures and generates new attack scenarios. The feedback unit can evaluate the effectiveness of the defensive measures and generate new attack scenarios based on the results. This enables the penetration testing support system to strengthen the security of the system from both the offensive and defensive perspectives.

[0099] 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 a voice indicating a user input for the result of the specific processing. The control unit 46A transmits the voice data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[0100] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search<URL: https: / / openai.com / blog / chatgpt> ) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating a voice, text data indicating a text, and image data indicating an image (e.g., still image data or video data) are input. The data generation model 58 infers the input inference data according to the instruction indicated by the prompt, and outputs the inference result in one or more data formats of voice data, text data, and image data. The data generation model 58 includes, for example, a text generation AI, an image generation AI, and a multimodal generation AI. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The specification processing unit 290 performs the above-mentioned specification processing while using the data generation model 58. The data generation model 58 may be a fine-tuned model so as to output an inference result from a prompt that does not include an instruction, and in this case, the data generation model 58 can output an inference result from a prompt that does not include an instruction. In the data processing device 12 and the like, the data generation model 58 includes a plurality of types, and the data generation model 58 includes an AI other than the generation AI. The AI ​​other than the generation AI is, for example, a linear regression, a logistic regression, a decision tree, a random forest, a support vector machine (SVM), a k-means clustering, a convolutional neural network (CNN), a recurrent neural network (RNN), a generative adversarial network (GAN), or a naive Bayes, and can perform various processes, but is not limited to such examples. The AI ​​may also be an AI agent. In addition, when the processes of each of the above-mentioned parts are performed by an AI, the processes are performed in part or in whole by the AI, but are not limited to such examples.In addition, processing performed by AI including the generation AI may be replaced with rule-based processing, and rule-based processing may be replaced with processing performed by AI including the generation AI.

[0101] Furthermore, the above-described processing by the data processing system 10 is executed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the smart device 14, but may also be executed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the smart device 14. Furthermore, the specific processing unit 290 of the data processing device 12 acquires or collects information required for processing from the smart device 14 or an external device, and the smart device 14 acquires or collects information required for processing from the data processing device 12 or an external device.

[0102] Each of the multiple elements including the attack unit, defense unit, and feedback unit described above is realized, for example, by at least one of the smart device 14 and the data processing device 12. For example, the attack unit is realized by the processor 46 of the smart device 14, and searches for vulnerabilities in the system and generates an attack scenario. The defense unit is realized, for example, by the specific processing unit 290 of the data processing device 12, and generates a defense measure based on the attack scenario and simulates its effect. The feedback unit is realized, for example, by the control unit 46A of the smart device 14, and evaluates the effect of the defense measure and generates a new attack scenario. The correspondence between each unit and the device or control unit is not limited to the above example, and various changes are possible.

[0103] [Second embodiment] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0104] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0105] 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. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN and / or a LAN.

[0106] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0107] The microphone 238 receives instructions and the like from the user by receiving voice uttered by the user. The microphone 238 captures the voice uttered by the user, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs the voice according to instructions from the processor 46.

[0108] Camera 42 is a small digital camera equipped with an optical system including a lens, an aperture, and a shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the user's surroundings (e.g., an imaging range defined by an angle of view equivalent to the width of the field of vision of an average healthy person).

[0109] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is performed in a secure state.

[0110] Fig. 4 shows an example of main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0111] The processor 28 reads the specific processing program 56 from the storage 32, and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0112] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290. The identification processing unit 290 can estimate the emotion of the user using the emotion identification model 59 and perform identification processing using the user's emotion. In the emotion estimation function (emotion identification function) using the emotion identification model 59, various estimations and predictions regarding the user's emotion, including estimation and prediction of the user's emotion, are performed, but are not limited to such examples. In addition, the estimation and prediction of the emotion also include, for example, analysis of the emotion.

[0113] In the smart glasses 214, the 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 operating as the control unit 46A in accordance with the specific processing program 60 executed on the RAM 48. Note that the smart glasses 214 have a data generation model and an emotion identification model similar to the data generation model 58 and the emotion identification model 59, and can also perform processing similar to that of the specific processing unit 290 using these models.

[0114] Note that a device other than the data processing device 12 may 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 a processing result (such as a prediction result) using the data generation model 58 by communicating with the server device having the data generation model 58. In addition, the data processing device 12 may be a server device, or may be a terminal device (such as a mobile phone, a robot, or a home appliance) owned by a user.

[0115] 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 a voice indicating a user input for the result of the specific processing. The control unit 46A transmits the voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[0116] The data generation model 58 is a so-called generative AI. An example of the 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. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating a voice, text data indicating a text, and image data indicating an image (e.g., still image data or video data) is input. The data generation model 58 infers the input inference data according to the instruction indicated by the prompt, and outputs the inference result in one or more data formats of voice data, text data, and image data. The data generation model 58 includes, for example, a text generation AI, an image generation AI, and a multimodal generation AI. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The specification processing unit 290 performs the above-mentioned specification processing while using the data generation model 58. The data generation model 58 may be a fine-tuned model so as to output an inference result from a prompt that does not include an instruction, and in this case, the data generation model 58 can output an inference result from a prompt that does not include an instruction. In the data processing device 12 and the like, the data generation model 58 includes a plurality of types, and the data generation model 58 includes an AI other than the generation AI. The AI ​​other than the generation AI is, for example, a linear regression, a logistic regression, a decision tree, a random forest, a support vector machine (SVM), a k-means clustering, a convolutional neural network (CNN), a recurrent neural network (RNN), a generative adversarial network (GAN), or a naive Bayes, and can perform various processes, but is not limited to such examples. The AI ​​may be an AI agent. In addition, when the processes of each part described above are performed by an AI, the processes are performed in part or in whole by the AI, but are not limited to such examples. In addition, the processes performed by the AI ​​including the generation AI may be replaced with rule-based processes, and the rule-based processes may be replaced with processes performed by the AI ​​including the generation AI.

[0117] 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 executed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the smart glasses 214, but may be executed 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 required for processing from the smart glasses 214 or an external device, and the smart glasses 214 acquires or collects information required for processing from the data processing device 12 or an external device.

[0118] Each of the multiple elements including the attack unit, defense unit, and feedback unit described above is realized, for example, by at least one of the smart glasses 214 and the data processing device 12. For example, the attack unit is realized by the processor 46 of the smart glasses 214, and searches for vulnerabilities in the system and generates an attack scenario. The defense unit is realized, for example, by the specific processing unit 290 of the data processing device 12, and generates a defense measure based on the attack scenario and simulates its effect. The feedback unit is realized, for example, by the control unit 46A of the smart glasses 214, and evaluates the effect of the defense measure and generates a new attack scenario. The correspondence between each unit and the device or control unit is not limited to the above example, and various changes are possible.

[0119] [Third embodiment] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0120] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0121] 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. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN and / or a LAN.

[0122] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0123] The microphone 238 receives instructions and the like from the user by receiving voice uttered by the user. The microphone 238 captures the voice uttered by the user, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs the voice according to instructions from the processor 46.

[0124] Camera 42 is a small digital camera equipped with an optical system including a lens, an aperture, and a shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the user's surroundings (e.g., an imaging range defined by an angle of view equivalent to the width of the field of vision of an average healthy person).

[0125] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is performed in a secure state.

[0126] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0127] The processor 28 reads the specific processing program 56 from the storage 32, and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0128] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290. The identification processing unit 290 can estimate the emotion of the user using the emotion identification model 59 and perform identification processing using the user's emotion. In the emotion estimation function (emotion identification function) using the emotion identification model 59, various estimations and predictions regarding the user's emotion, including estimation and prediction of the user's emotion, are performed, but are not limited to such examples. In addition, the estimation and prediction of the emotion also include, for example, analysis of the emotion.

[0129] In the headset type terminal 314, the 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 operating as the control unit 46A in accordance with the specific program 60 executed on the RAM 48. Note that the headset type terminal 314 has a data generation model and an emotion identification model similar to the data generation model 58 and the emotion identification model 59, and can also perform processing similar to that of the identification processing unit 290 using these models.

[0130] Note that a device other than the data processing device 12 may 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 a processing result (such as a prediction result) using the data generation model 58 by communicating with the server device having the data generation model 58. In addition, the data processing device 12 may be a server device, or may be a terminal device (such as a mobile phone, a robot, or a home appliance) owned by a user.

[0131] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input for the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[0132] The data generation model 58 is a so-called generative AI. An example of the 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. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating a voice, text data indicating a text, and image data indicating an image (e.g., still image data or video data) is input. The data generation model 58 infers the input inference data according to the instruction indicated by the prompt, and outputs the inference result in one or more data formats of voice data, text data, and image data. The data generation model 58 includes, for example, a text generation AI, an image generation AI, and a multimodal generation AI. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The specification processing unit 290 performs the above-mentioned specification processing while using the data generation model 58. The data generation model 58 may be a fine-tuned model so as to output an inference result from a prompt that does not include an instruction, and in this case, the data generation model 58 can output an inference result from a prompt that does not include an instruction. In the data processing device 12 and the like, the data generation model 58 includes a plurality of types, and the data generation model 58 includes an AI other than the generation AI. The AI ​​other than the generation AI is, for example, a linear regression, a logistic regression, a decision tree, a random forest, a support vector machine (SVM), a k-means clustering, a convolutional neural network (CNN), a recurrent neural network (RNN), a generative adversarial network (GAN), or a naive Bayes, and can perform various processes, but is not limited to such examples. The AI ​​may be an AI agent. In addition, when the processes of each part described above are performed by an AI, the processes are performed in part or in whole by the AI, but are not limited to such examples. In addition, the processes performed by the AI ​​including the generation AI may be replaced with rule-based processes, and the rule-based processes may be replaced with processes performed by the AI ​​including the generation AI.

[0133] 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 executed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the headset type terminal 314, but may be executed by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the headset type terminal 314. In addition, the specific processing unit 290 of the data processing device 12 acquires or collects information required for processing from the headset type terminal 314 or an external device, and the headset type terminal 314 acquires or collects information required for processing from the data processing device 12 or an external device.

[0134] Each of the multiple elements including the attack unit, defense unit, and feedback unit described above is realized, for example, by at least one of the headset type terminal 314 and the data processing device 12. For example, the attack unit is realized by the processor 46 of the headset type terminal 314, and searches for vulnerabilities in the system and generates an attack scenario. The defense unit is realized, for example, by the specific processing unit 290 of the data processing device 12, and generates defense measures based on the attack scenario and simulates their effects. The feedback unit is realized, for example, by the control unit 46A of the headset type terminal 314, and evaluates the effects of the defense measures and generates a new attack scenario. The correspondence between each unit and the device or control unit is not limited to the above example, and various changes are possible.

[0135] [Fourth embodiment] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[0136] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[0137] 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. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN and / or a LAN.

[0138] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[0139] The microphone 238 receives instructions and the like from the user by receiving voice uttered by the user. The microphone 238 captures the voice uttered by the user, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs the voice according to instructions from the processor 46.

[0140] Camera 42 is a small digital camera equipped with an optical system including a lens, an aperture, and a shutter, and an imaging element such as a CMOS image sensor or a CCD image sensor, and captures images of the user's surroundings (e.g., an imaging range defined by an angle of view equivalent to the width of the field of vision of a typical healthy person).

[0141] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is performed in a secure state.

[0142] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and legs. The posture and behavior of the robot 414 are controlled by controlling the motors of the arms, hands, and legs. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[0143] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0144] The processor 28 reads the specific processing program 56 from the storage 32, and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0145] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290. The identification processing unit 290 can estimate the emotion of the user using the emotion identification model 59 and perform identification processing using the user's emotion. In the emotion estimation function (emotion identification function) using the emotion identification model 59, various estimations and predictions regarding the user's emotion, including estimation and prediction of the user's emotion, are performed, but are not limited to such examples. In addition, the estimation and prediction of the emotion also include, for example, analysis of the emotion.

[0146] In the robot 414, the specific processing is performed by the processor 46. The storage 50 stores a specific program 60. The processor 46 reads out the specific program 60 from the storage 50, and executes the specific program 60 on the RAM 48. The specific processing is realized by the processor 46 operating as the control unit 46A in accordance with the specific program 60 executed on the RAM 48. The robot 414 has a data generation model and an emotion identification model similar to the data generation model 58 and the emotion identification model 59, and can also perform processing similar to that of the specific processing unit 290 using these models.

[0147] Note that a device other than the data processing device 12 may 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 a processing result (such as a prediction result) using the data generation model 58 by communicating with the server device having the data generation model 58. In addition, the data processing device 12 may be a server device, or may be a terminal device (such as a mobile phone, a robot, or a home appliance) owned by a user.

[0148] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires a voice indicating a user input for the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[0149] The data generation model 58 is a so-called generative AI. An example of the 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. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating a voice, text data indicating a text, and image data indicating an image (e.g., still image data or video data) is input. The data generation model 58 infers the input inference data according to the instruction indicated by the prompt, and outputs the inference result in one or more data formats of voice data, text data, and image data. The data generation model 58 includes, for example, a text generation AI, an image generation AI, and a multimodal generation AI. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The specification processing unit 290 performs the above-mentioned specification processing while using the data generation model 58. The data generation model 58 may be a fine-tuned model so as to output an inference result from a prompt that does not include an instruction, and in this case, the data generation model 58 can output an inference result from a prompt that does not include an instruction. In the data processing device 12 and the like, the data generation model 58 includes a plurality of types, and the data generation model 58 includes an AI other than the generation AI. The AI ​​other than the generation AI is, for example, a linear regression, a logistic regression, a decision tree, a random forest, a support vector machine (SVM), a k-means clustering, a convolutional neural network (CNN), a recurrent neural network (RNN), a generative adversarial network (GAN), or a naive Bayes, and can perform various processes, but is not limited to such examples. The AI ​​may be an AI agent. In addition, when the processes of each part described above are performed by an AI, the processes are performed in part or in whole by the AI, but are not limited to such examples. In addition, the processes performed by the AI ​​including the generation AI may be replaced with rule-based processes, and the rule-based processes may be replaced with processes performed by the AI ​​including the generation AI.

[0150] 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 executed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the robot 414, but may be executed 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 required for processing from the robot 414 or an external device, etc., and the robot 414 acquires or collects information required for processing from the data processing device 12 or an external device, etc.

[0151] Each of the multiple elements including the attack unit, defense unit, and feedback unit described above is realized, for example, by at least one of the robot 414 and the data processing device 12. For example, the attack unit is realized by the processor 46 of the robot 414, and searches for vulnerabilities in the system and generates an attack scenario. The defense unit is realized, for example, by the specific processing unit 290 of the data processing device 12, and generates a defense measure based on the attack scenario and simulates its effect. The feedback unit is realized, for example, by the control unit 46A of the robot 414, and evaluates the effect of the defense measure and generates a new attack scenario. The correspondence between each unit and the device or control unit is not limited to the above example, and various changes are possible.

[0152] The emotion identification model 59 as an emotion engine may determine the emotion of the user according to a specific mapping. Specifically, the emotion identification model 59 may determine the emotion of the user according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the emotion of the robot, and the identification processing unit 290 may perform identification processing using the emotion of the robot.

[0153] FIG. 9 is a diagram showing an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive emotions are arranged. The more outside the concentric circles, the more emotions that represent states and actions that arise from a state of mind are arranged. Emotions are a concept that includes emotions and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions that occur in the brain are arranged. On the right side of the concentric circles, emotions that are generally induced by situational judgment are arranged. On the upper and lower sides of the concentric circles, emotions that are generally generated from reactions that occur in the brain and are induced by situational judgment are arranged. In addition, on the upper side of the concentric circles, emotions of "pleasure" are arranged, and on the lower side, emotions of "discomfort" are arranged. In this way, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[0154] These emotions are distributed in the three o'clock direction of emotion map 400 and usually fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[0155] The inside of emotion map 400 represents what is going on inside one's mind, and the outside of emotion map 400 represents behavior, so the further out on emotion map 400 you go, the more visible the emotions become (the more they are expressed in behavior).

[0156] Here, human emotions are based on various balances such as posture and blood sugar level, and when these balances are far from the ideal, it indicates an unpleasant state, and when they are close to the ideal, it indicates a pleasant state. Emotions can also be created for robots, cars, motorcycles, etc., based on various balances such as posture and battery level, so that when these balances are far from the ideal, it indicates an unpleasant state, and when they are close to the ideal, it indicates a pleasant state. The emotion map may be generated, for example, based on the emotion map of Dr. Mitsuyoshi (Research on speech emotion recognition and emotion brain physiological signal analysis system, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). On the left half of the emotion map, emotions belonging to an area called "reaction" where sensation is dominant are lined up. On the right half of the emotion map, emotions belonging to an area called "situation" where situation recognition is dominant are lined up.

[0157] The emotion map defines two emotions that promote learning. The first is the negative emotion around the middle of "repentance" or "remorse" on the situation side. In other words, this 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 positive emotion around "desire" on the response side. In other words, this is when the robot has positive feelings such as "I want more" or "I want to know more."

[0158] The emotion identification model 59 inputs the user input to a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the emotion of the user. This neural network is pre-trained based on multiple learning data that are combinations of the user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in Fig. 10. Fig. 10 shows an example in which multiple emotions, "relief," "calm," and "encouraging," have similar emotion values.

[0159] In the above embodiment, an example was given in which a specific process is performed by one computer 22, but the technology disclosed herein is not limited to this, and distributed processing of the specific process may be performed by multiple computers, including computer 22.

[0160] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable computer-readable non-transitory storage medium such as a Universal Serial Bus (USB) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[0161] In addition, 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 upon request from the data processing device 12.

[0162] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[0163] As the hardware resource for executing the specific process, various processors as shown below can be used. An example of the processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing the specific process by executing software, i.e., a program. Another example of the processor is a dedicated electric circuit, which is a processor having a circuit configuration designed exclusively for executing a specific process, such as a Field-Programmable Gate Array (FPGA), a Programmable Logic Device (PLD), or an Application Specific Integrated Circuit (ASIC). Each processor has a built-in or connected memory, and each processor executes the specific process by using the memory.

[0164] The hardware resource that executes the specific process may be one of these various processors, or may be a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific process may be a single processor.

[0165] As an example of a configuration using one processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a configuration using a processor that realizes the functions of the entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[0166] Furthermore, more specifically, the hardware structure of these various processors can be an electric circuit that combines circuit elements such as semiconductor elements. The specific processes described above are merely examples. It goes without saying that unnecessary steps may be deleted, new steps may be added, or the order of processes may be changed without departing from the spirit of the invention.

[0167] In the above example, the first to fourth embodiments are described separately, but some or all of these embodiments may be combined. The smart device 14, the smart glasses 214, the headset terminal 314, and the robot 414 are only examples, and they may be combined with each other, or may be other devices. In the above example, the first and second embodiments are described separately, but they may be combined.

[0168] The above description and illustrations are detailed descriptions of the parts related to the technology of the present disclosure, and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, function, action, and effect is an example of the configuration, function, action, and effect of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replaced with respect to the description and illustrations shown above, within the scope of the gist of the technology of the present disclosure. In addition, in order to avoid confusion and to facilitate understanding of the parts related to the technology of the present disclosure, the description and illustrations shown above omit explanations of technical common sense that do not require particular explanation in enabling the implementation of the technology of the present disclosure.

[0169] All publications, patent applications, and standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, and standard was specifically and individually indicated to be incorporated by reference.

[0170] (Appendix 1) An attack team that searches for vulnerabilities in the system, a defense unit that generates a defense measure based on the attack scenario generated by the attack unit; a feedback unit that evaluates the effectiveness of the defense measures generated by the defense unit and generates a new attack scenario. A system characterized by: (Appendix 2) The attacking section includes: Deeply analyze each component of your system to identify potential vulnerabilities 2. The system of claim 1, (Appendix 3) The attacking section includes: Generate attack scenarios based on identified vulnerabilities 2. The system of claim 1, (Appendix 4) The defense unit is Analyze attack scenarios and propose appropriate defense measures 2. The system of claim 1, (Appendix 5) The defense unit is Simulate the effectiveness of proposed defense measures and evaluate their actual effectiveness 2. The system of claim 1, (Appendix 6) The feedback unit is Evaluate the effectiveness of defense measures and generate new attack scenarios 2. The system of claim 1, (Appendix 7) The attacking section includes: Infer user sentiment and adjust vulnerability search priorities based on the inferred user sentiment 2. The system of claim 1, (Appendix 8) The attacking section includes: Analyze the system's past attack history and select the most effective vulnerability detection method 2. The system of claim 1, (Appendix 9) The attacking section includes: When searching for vulnerabilities, adjust the search method based on the system's operating status and load. 2. The system of claim 1, (Appendix 10) The attacking section includes: Infer user sentiment and adjust vulnerability reporting based on the inferred user sentiment 2. The system of claim 1, (Appendix 11) The attacking section includes: When hunting for vulnerabilities, adjust your hunting techniques to take into account the geographic distribution of systems. 2. The system of claim 1, (Appendix 12) The attacking section includes: When searching for vulnerabilities, improve the accuracy of the search by referring to the system's related literature. 2. The system of claim 1, (Appendix 13) The defense unit is Inferring user emotions and adjusting the method of proposing defense measures based on the estimated user emotions 2. The system of claim 1, (Appendix 14) The defense unit is When generating defense measures, the most appropriate defense measures are selected by referring to past defense history. 2. The system of claim 1, (Appendix 15) The defense unit is Customize defenses based on the system's current security policy when generating them 2. The system of claim 1, (Appendix 16) The defense unit is Inferring user sentiment and prioritizing defense measures based on the inferred user sentiment 2. The system of claim 1, (Appendix 17) The defense unit is When generating defense measures, optimize them by taking into account the geographical distribution of systems. 2. The system of claim 1, (Appendix 18) The defense unit is When generating defense measures, the system refers to related literature to improve the accuracy of the defense measures. 2. The system of claim 1, (Appendix 19) The feedback unit is Estimate user's emotions and adjust the feedback method based on the estimated user's emotions 2. The system of claim 1, (Appendix 20) The feedback unit is When providing feedback, select the most appropriate feedback method by referring to past feedback history 2. The system of claim 1, (Appendix 21) The feedback unit is When providing feedback, customize the feedback content based on the current state of the system 2. The system of claim 1, (Appendix 22) The feedback unit is Estimate user emotions and prioritize feedback based on the estimated user emotions 2. The system of claim 1, (Appendix 23) The feedback unit is During feedback, optimize the feedback method by taking into account the geographical distribution of systems. 2. The system of claim 1, (Appendix 24) The feedback unit is When providing feedback, refer to the relevant literature of the system to improve the accuracy of the feedback. 2. The system of claim 1, [Explanation of symbols]

[0171] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot

Claims

1. An attack team that searches for vulnerabilities in the system, a defense unit that generates a defense measure based on the attack scenario generated by the attack unit; a feedback unit that evaluates the effectiveness of the defense measures generated by the defense unit and generates a new attack scenario. A system characterized by:

2. The attacking section includes: Deeply analyze each component of your system to identify potential vulnerabilities 2. The system of claim 1 .

3. The attacking section includes: Generate attack scenarios based on identified vulnerabilities 2. The system of claim 1 .

4. The defense unit is Simulate the effectiveness of proposed defense measures and evaluate their actual effectiveness 2. The system of claim 1 .

5. The feedback unit is Evaluate the effectiveness of defense measures and generate new attack scenarios 2. The system of claim 1 .

6. The attacking section includes: Infer user sentiment and adjust vulnerability search priorities based on the inferred user sentiment 2. The system of claim 1 .

7. The attacking section includes: Analyze the system's past attack history and select the most effective vulnerability detection method 2. The system of claim 1 .

Images