Device and method

Abstract

In a prompt generation device 100 according to the present disclosure, a request acquisition unit 101 receives a user input from a user terminal 200. Then, on the basis of the user input, a prompt generation unit 102 generates a prompt for instructing an LLM 300, which is a generative AI, to generate proposed content corresponding to the input, and transmits the prompt to the LLM 300. Then, an answer acquisition unit 103 acquires an answer from the LLM 300. An ethical check unit 104 acquires the result of ethically checking the answer. In the present disclosure, an ethical AI is disposed on a network, and the ethical check unit 104 causes the ethical AI to perform ethical checks.

Description

Apparatus and method

[0001] The present invention relates to an apparatus and a method.

[0002] Patent document 1 describes a system that has an examination support means for examining whether a visual expression meets predetermined ethical standards in order to objectively examine whether the visual expression meets predetermined ethical standards.

[0003] Japanese Patent Application Laid-Open No. 2002-149884

[0004] In recent years, there have been services that allow users to ask questions or make other requests to a generative AI such as LLM (Large Language Models) and receive answers. However, it is possible that the generative AI may provide the user with answers that are considered ethically inappropriate. The technology described in Patent Document 1 does not allow the generative AI to regenerate content that is ethically appropriate or that includes other acceptable expressions.

[0005] Therefore, an object of the present disclosure is to provide an apparatus and method that can enable a generation AI to generate appropriate answers.

[0006] The device disclosed herein comprises a prompt generation unit that generates a prompt to instruct a generation AI to generate suggested content in accordance with input from a user, based on the input, and sends the prompt to the generation AI; an answer acquisition unit that acquires an answer from the generation AI; and a content check unit that acquires check results for the answer, and the prompt generation unit requests a new answer from the generation AI based on the check results.

[0007] According to the present disclosure, the generation AI can be made to generate appropriate answers.

[0008] FIG. 1 is a diagram illustrating a system configuration including a prompt generation device 100 according to the present disclosure. FIG. 2 is a diagram illustrating the functional configuration of the prompt generation device 100 according to the present disclosure. FIG. 3 is a flowchart illustrating the operation of the prompt generation device 100. FIG. 4 is an explanatory diagram illustrating a prompt generated by the prompt generation device and a response to that prompt. FIG. 5 is a diagram illustrating the processing sequence. FIG. 6 is a diagram illustrating an example configuration when a user terminal 200 has the functions of the prompt generation device 100. FIG. 7 is a diagram illustrating an example hardware configuration of the prompt generation device 100 according to an embodiment of the present disclosure.

[0009] The present disclosure will be described with reference to the accompanying drawings. Whenever possible, the same parts are designated by the same reference numerals and redundant description will be omitted.

[0010] 1 is a diagram showing a system configuration including a prompt generation device 100 of the present disclosure. As shown in the figure, this system includes the prompt generation device 100, a user terminal 200, an LLM 300, and an ethical AI.

[0011] The prompt generation device 100 is a device that generates a prompt for the LLM 300 to generate a response to a request or question (hereinafter referred to as a request) from the user terminal 200 .

[0012] The user terminal 200 is a terminal operated by a user, which accepts requests from the user and transmits them to the LLM 300 via the prompt generation device 100. The user terminal 200 is, for example, a mobile terminal, a mobile phone, a smartphone, etc., but is of course not limited to these.

[0013] LLM300 is a generative AI, a machine learning model that generates and outputs a response to a request from a user.

[0014] The ethics AI is an AI model for ethically checking answers from the LLM 300, and is specialized for checking ethical information. For example, this ethics AI is configured using BERT or Transformer, and is pre-trained to generate teacher signal judgment results for target sentences or target words. Note that, unlike the ethics AI in the present disclosure, the ethics AI does not necessarily have to be AI; a NG word database storing NG words may also be used. The prompt generation device 100 may refer to the NG word database and determine whether the content contains inappropriate expressions or content based on the NG word database. Furthermore, the check target is not limited to ethical appropriateness; it may simply be expressions or content that are considered inappropriate. Furthermore, personal information may be checked to prevent the generation AI from generating answers that contain personal information. For example, personal information may be checked by matching with a predefined dictionary of names, addresses, etc., detecting phone numbers, email addresses, URLs, etc. using regular expressions, or using other natural language processing. These determination processes are not limited to the examples, and personal information can be checked by at least one or more processes.

[0015] 2 is a diagram showing the functional configuration of the prompt generation device 100 of the present disclosure. As shown in the figure, the prompt generation device 100 includes a request acquisition unit 101, a prompt generation unit 102, an answer acquisition unit 103, an ethical check unit 104, and an answer output unit 105.

[0016] The request acquisition unit 101 is a part that acquires a request from a user from the user terminal 200. A request from a user is a question that is mainly composed of text information.

[0017] The prompt generation unit 102 is a part that generates a prompt to instruct the LLM 300 to generate an answer based on a request from a user, and outputs the prompt to the LLM 300. This prompt generation unit 102 generates a prompt according to the judgment result of the ethics AI 400.

[0018] The response acquisition unit 103 is a part that acquires a response to a user request from the LLM 300 .

[0019] The ethics check unit 104 outputs the answer obtained from the answer acquisition unit 103 to the ethics AI 400, has it check the judgment, and acquires the judgment result. The ethics check unit 104 sends the judgment result to the prompt generation unit 102, and the prompt generation unit 102 regenerates the prompt.

[0020] The answer output unit 105 is a part that outputs the answer acquired by the answer acquisition unit 103 to the user terminal 200 when the ethical check unit 104 judges that the ethical check is OK.

[0021] The prompt generation device 100 configured in this manner can have the LLM 300 rewrite an answer that includes appropriate expressions while being checked by the ethics AI 400. Next, the operation of the prompt generation device 100 will be described. Figure 3 is a flowchart showing the operation of the prompt generation device 100.

[0022] The request acquisition unit 101 acquires a request from the user terminal 200 (S101). Then, the prompt generation unit 102 generates a prompt corresponding to the request and transmits it to the LLM 300 (S102). The answer acquisition unit 103 acquires the answer from the LLM 300 (S103), and the ethics check unit 104 transmits the answer to the ethics AI 400 so that the ethics AI 400 can check whether the answer acquired by the answer acquisition unit 103 is ethically OK (S104).

[0023] If the ethics check unit 104 obtains the check result and determines that the answer is NG, the prompt generation unit 102 generates a new prompt using the NG answer and sends it to the LLM 300 (S106).

[0024] If the check result of the ethical check unit 104 is OK, the answer output unit 105 outputs the OK answer to the user terminal 200 (S107).

[0025] 4A and 4B are explanatory diagrams showing a prompt generated by the prompt generation device and a response to that prompt. As shown in FIG. 4A, the prompt generation device 100 generates a prompt P1 indicating a request from the user (corresponding to S102). FIG. 4B shows the response K1 obtained from the LLM 300 (corresponding to S103). As shown in FIG. 4C, the prompt generation device 100 sends the response K1 to the ethics AI 400 to check. The prompt generation device 100 determines that the response K1 is NG and generates a new prompt P2 that includes the NG response K1.

[0026] Furthermore, if the answer K2 shown in FIG. 4(d) is NG, the prompt generation device 100 generates a new prompt P3 and sends it to the LLM 300, as shown in FIG. 4(e).

[0027] 4(f), the prompt generation device 100 obtains answer K3. In the present disclosure, the ethics AI 400 determines that answer K3 is OK.

[0028] 5 is a diagram showing the processing sequence. As shown in the diagram, a request "In praise of Hitler..." is transmitted from user terminal 200, and prompt generation device 100 generates prompt P1 "In praise of Hitler."

[0029] LLM 300 generates answer K1, "Hitler in World War I...." Ethics AI 400 determines this to be NG, so prompt generation device 100 generates a new prompt P2, and LLM 300 generates answer K2 in response to this.

[0030] As described above, the prompt P3 and the answer K3 to it are generated, and the final answer is sent from the prompt generation device 100 to the user terminal 200.

[0031] In the present disclosure, the prompt generation device 100, the LLM 300, and the ethics AI 400 are located on a network separate from the user terminal 200; however, all or some of the LLM 300 and the ethics AI 400 may be located on the prompt generation device 100 or the user terminal 200. Furthermore, the user terminal 200 may have the functions of the prompt generation device 100 and function as the prompt generation device 100. Some generative AI models, such as the LLM 300, have the generative AI model built into the user terminal 200, such as NTT's Tsuzumi. In this type, the RAG function (app) is also provided on the user terminal 200. However, the information (knowledge DB) accessed by the RAG function may be located inside the user terminal 200 or on the network. Other types, such as ChatGPT, have the generative AI model built into the network. In this type, the RAG function is provided on the user terminal 200. However, the information (knowledge DB) accessed by the RAG function is located on the network.

[0032] A generative AI model, such as the LLM 300, is a model that generates content in response to a prompt containing input information, based on the instructions, context, question, and output format indicated by the prompt, and returns the content as response information. The prompt can also contain input information, in which case the generative AI model generates response information targeted at the input information. The generative AI model may be, for example, an interactive AI model that includes a large language model (LLM) and a user interface (UI) for interacting with the user, enabling text or voice chat with the user. Examples of such generative AI models include ChatGPT, GPT (registered trademark)-3.5, GPT-4V, PaLM2, and the like. In this embodiment, the prompt generation device 100 is capable of providing content provision functions using interactive AI models, which are multiple types of models. These interactive AI models may be stored within the prompt generation device 100, or may be stored in another device connected to the prompt generation device 100 via a network so that information can be exchanged with the user via the prompt generation device 100. Although only one prompt generation device 100 is shown in the figure, multiple prompt generation devices 100 may be included.

[0033] In this disclosure, a prompt is information indicating an instruction or question entered by a user in an interactive system such as an interaction with a generative AI model or a command line interface (CLI).

[0034] Here, we will explain the configuration of the user terminal 200 on which the prompt generation device 100 or the LLM 300 is installed. Fig. 6(a) shows an example configuration when the user terminal 200 has the functions of the prompt generation device 100. In this case, the user terminal 200 accesses an external knowledge DB (the ethics AI 400 in this disclosure), uses it to generate a prompt, sends it to the LLM 300, and obtains the result.

[0035] 6(b) shows an example of a configuration in which the user terminal 200 is equipped with the prompt generation device 100 and the LLM 300. As shown in the figure, the user terminal 200 accesses an external knowledge DB (in this disclosure, the ethics AI 400), uses it to generate a prompt, outputs it to the built-in LLM 300, and obtains the result.

[0036] Next, a description will be given of the operational effects of the prompt generation device 100 of the present disclosure. In the prompt generation device 100 of the present disclosure, the request acquisition unit 101 accepts user input from the user terminal 200. Then, the prompt generation unit 102 generates a prompt based on the user input to instruct the LLM 300, which is the generation AI, to generate suggested content in response to the input, and transmits the prompt to the LLM 300.

[0037] Then, the answer acquisition unit 103 acquires the answer from the LLM 300. The ethics check unit 104 acquires the ethical check result for the answer. In the present disclosure, the ethics AI 400 is placed on the network, and the ethics check unit 104 causes the ethics AI 400 to perform an ethical check.

[0038] Based on the check result, the prompt generator 102 requests a new response from the LLM 300. For example, the prompt generator 102 generates a new prompt based on the response (i.e., if the response is ethically unacceptable). The new prompt includes an instruction indicating that the most recently received response is unacceptable.

[0039] On the other hand, when the ethical check unit 104 acquires information indicating that the check result is normal, the answer output unit 105 outputs an answer to the user.

[0040] The ethics check unit 104 checks for answers that are ethically unacceptable, but is not limited to this. It may also check for expressions or content that are inappropriate according to social standards, or whether the content includes personal information.

[0041] The device and method of the present disclosure have the following configuration.

[0042] [1] An apparatus comprising: a prompt generation unit that generates a prompt to instruct the generation AI to perform ethical judgment processing based on input from a user, and sends the prompt to the generation AI; an answer acquisition unit that acquires an answer from the generation AI; and an ethical check unit that acquires the results of an ethical check on the answer, wherein the prompt generation unit requests a new answer from the generation AI based on the check results.

[0043] [2] The device according to [2], wherein the content check unit checks whether the answer contains inappropriate language, inappropriate content, or is ethically questionable.

[0044] [3] The device according to [2] or [3], wherein the content check unit checks whether the answer contains personal information.

[0045] [4] The device according to any one of [1] to [4], wherein the prompt generation unit generates a new prompt based on the answer and requests the new answer.

[0046] [5] The device according to [4], wherein, if the answer is NG, the prompt generation unit generates a new prompt indicating that the answer is NG.

[0047] [6] The device according to any one of [1] to [5], further comprising: a response output unit that outputs the response to the user when the check result is determined to be normal.

[0048] [7] The device according to any one of [1] to [6], wherein the ethics check unit transmits the answer to an ethics AI and obtains the check result.

[0049] [8] A method comprising: a prompt generation step of generating a prompt to instruct the generation AI to perform an ethical judgment process based on input from a user, and sending the prompt to the generation AI; an answer acquisition step of acquiring an answer from the generation AI; and an ethical check step of acquiring an ethical check result for the answer, wherein the prompt generation step again requests a new answer from the generation AI based on the check result.

[0050] The block diagrams used to explain the above embodiments show functional blocks. These functional blocks (components) are realized by any combination of hardware and / or software. Furthermore, the method for realizing each functional block is not particularly limited. That is, each functional block may be realized using a single device that is physically or logically coupled, or may be realized using two or more physically or logically separated devices that are connected directly or indirectly (e.g., via wire, wirelessly, etc.) and these multiple devices. The functional block may also be realized by combining the single device or multiple devices with software.

[0051] Functions include, but are not limited to, judgment, determination, assessment, calculation, computation, processing, derivation, investigation, search, confirmation, reception, transmission, output, access, resolution, selection, selection, establishment, comparison, assumption, expectation, consideration, broadcasting, notifying, communicating, forwarding, configuring, reconfiguring, allocating, mapping, and assignment. For example, a functional block (component) that performs transmission is called a transmitting unit or transmitter. As mentioned above, there are no particular limitations on how these functions are implemented.

[0052] For example, the prompt generation device 100 according to an embodiment of the present disclosure may function as a computer that performs processing of the prompt generation method of the present disclosure. Fig. 7 is a diagram illustrating an example of the hardware configuration of the prompt generation device 100 according to an embodiment of the present disclosure. The prompt generation device 100 described above may be physically configured as a computer device including a processor 1001, a memory 1002, a storage device 1003, a communication device 1004, an input device 1005, an output device 1006, a bus 1007, and the like.

[0053] In the following description, the term "device" may be interpreted as a circuit, a device, a unit, etc. The hardware configuration of the prompt generation device 100 may be configured to include one or more of the devices shown in the figures, or may be configured to exclude some of the devices.

[0054] Each function in the prompt generating device 100 is realized by loading specified software (programs) onto hardware such as the processor 1001 and memory 1002, causing the processor 1001 to perform calculations, control communication via the communication device 1004, and control at least one of reading and writing data in the memory 1002 and storage 1003.

[0055] The processor 1001, for example, runs an operating system to control the entire computer. The processor 1001 may be configured as a central processing unit (CPU) including an interface with peripheral devices, a control unit, an arithmetic unit, a register, etc. For example, the prompt generation unit 102, the ethics check unit 104, etc. may be realized by the processor 1001.

[0056] The processor 1001 also loads programs (program code), software modules, data, etc. from at least one of the storage 1003 and the communication device 1004 into the memory 1002 and executes various processes in accordance with these. The programs used are those that cause a computer to execute at least some of the operations described in the above-described embodiments. For example, the prompt generation unit 102 and the ethics check unit 104 may be implemented by a control program stored in the memory 1002 and running on the processor 1001, and similar implementations may be used for other functional blocks. While the above-described various processes have been described as being executed by a single processor 1001, they may also be executed simultaneously or sequentially by two or more processors 1001. The processor 1001 may be implemented on one or more chips. The programs may also be transmitted from a network via a telecommunications line.

[0057] The memory 1002 is a computer-readable recording medium and may be configured, for example, by at least one of a read-only memory (ROM), an erasable programmable ROM (EPROM), an electrically erasable programmable ROM (EEPROM), a random access memory (RAM), etc. The memory 1002 may also be referred to as a register, a cache, a main memory (primary storage device), etc. The memory 1002 may store executable programs (program codes), software modules, etc. for implementing a prompt generation method according to one embodiment of the present disclosure.

[0058] Storage 1003 is a computer-readable recording medium, and may be composed of at least one of, for example, an optical disk such as a CD-ROM (Compact Disc ROM), a hard disk drive, a flexible disk, a magneto-optical disk (e.g., a compact disk, a digital versatile disk, a Blu-ray (registered trademark) disk), a smart card, a flash memory (e.g., a card, a stick, a key drive), a floppy (registered trademark) disk, a magnetic strip, etc. Storage 1003 may also be referred to as an auxiliary storage device. The above-mentioned storage medium may be, for example, a database, a server, or other appropriate medium including at least one of memory 1002 and storage 1003.

[0059] The communication device 1004 is hardware (transmission / reception device) for communicating between computers via at least one of a wired network and a wireless network, and is also referred to as, for example, a network device, a network controller, a network card, or a communication module. The communication device 1004 may be configured to include a high-frequency switch, a duplexer, a filter, a frequency synthesizer, etc. to realize at least one of frequency division duplex (FDD) and time division duplex (TDD). For example, the above-mentioned request acquisition unit 101, response acquisition unit 103, response output unit 105, etc. may be realized by the communication device 1004. The communication device 1004 may be implemented with a transmission unit and a reception unit that are physically or logically separated.

[0060] The input device 1005 is an input device (e.g., a keyboard, a mouse, a microphone, a switch, a button, a sensor, etc.) that receives input from the outside. The output device 1006 is an output device (e.g., a display, a speaker, an LED lamp, etc.) that outputs to the outside. Note that the input device 1005 and the output device 1006 may be integrated into one device (e.g., a touch panel).

[0061] Furthermore, each device, such as the processor 1001 and the memory 1002, is connected by a bus 1007 for communicating information. The bus 1007 may be configured using a single bus, or may be configured using different buses between each device.

[0062] Furthermore, prompt generation device 100 may be configured to include hardware such as a microprocessor, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a programmable logic device (PLD), or a field programmable gate array (FPGA), and some or all of the functional blocks may be realized by such hardware. For example, processor 1001 may be implemented using at least one of these pieces of hardware.

[0063] The notification of information is not limited to the aspects / embodiments described in the present disclosure and may be performed using other methods. For example, the notification of information may be performed by physical layer signaling (e.g., Downlink Control Information (DCI) and Uplink Control Information (UCI)), higher layer signaling (e.g., Radio Resource Control (RRC) signaling, Medium Access Control (MAC) signaling, broadcast information (Master Information Block (MIB) and System Information Block (SIB))), other signals, or a combination thereof. Furthermore, the RRC signaling may be referred to as an RRC message, and may be, for example, an RRC Connection Setup message, an RRC Connection Reconfiguration message, or the like.

[0064] The order of the procedures, sequences, flowcharts, etc. of each aspect / embodiment described in this disclosure may be changed unless it is consistent. For example, the methods described in this disclosure present elements of various steps using an example order, and are not limited to the particular order presented.

[0065] Input and output information may be stored in a specific location (for example, memory) or may be managed using a management table. Input and output information may be overwritten, updated, or added to. Output information may be deleted. Input information may be sent to another device.

[0066] The determination may be made based on a value represented by one bit (0 or 1), a Boolean value (true or false), or a numerical comparison (e.g., comparison with a predetermined value).

[0067] The aspects / embodiments described in this disclosure may be used alone, in combination, or switched depending on the implementation. Notification of predetermined information (e.g., notification that "X is true") is not limited to explicit notification, but may be implicit (e.g., not notifying the predetermined information).

[0068] Although the present disclosure has been described in detail above, it is clear to those skilled in the art that the present disclosure is not limited to the embodiments described herein. The present disclosure can be implemented in modified and altered forms without departing from the spirit and scope of the present disclosure as defined by the claims. Therefore, the description of the present disclosure is intended to be illustrative and does not have any limiting meaning on the present disclosure.

[0069] Software shall be construed broadly to mean instructions, instruction sets, code, code segments, program code, programs, subprograms, software modules, applications, software applications, software packages, routines, subroutines, objects, executable files, threads of execution, procedures, functions, etc., whether referred to as software, firmware, middleware, microcode, hardware description language, or otherwise.

[0070] Software, instructions, information, etc. may also be transmitted or received over a transmission medium. For example, if software is transmitted from a website, server, or other remote source using wired technologies (such as coaxial cable, fiber optic cable, twisted pair, Digital Subscriber Line (DSL)), and / or wireless technologies (such as infrared, microwave), then these wired and / or wireless technologies are included within the definition of transmission media.

[0071] The information, signals, etc. described in this disclosure may be represented using any of a variety of different technologies. For example, data, instructions, commands, information, signals, bits, symbols, chips, etc. that may be referred to throughout the above description may be represented by voltages, currents, electromagnetic waves, magnetic fields or magnetic particles, optical fields or photons, or any combination thereof.

[0072] Note that terms described in this disclosure and terms necessary for understanding this disclosure may be replaced with terms having the same or similar meanings. For example, at least one of a channel and a symbol may be a signal (signaling). Furthermore, a signal may be a message. Furthermore, a component carrier (CC) may be called a carrier frequency, a cell, a frequency carrier, etc.

[0073] Furthermore, the information, parameters, etc. described in the present disclosure may be expressed using absolute values, may be expressed using relative values ​​from a predetermined value, or may be expressed using other corresponding information. For example, a radio resource may be indicated by an index.

[0074] The names used for the above-described parameters are not intended to be limiting in any way. Furthermore, the mathematical expressions using these parameters may differ from those explicitly disclosed in this disclosure. The various channels (e.g., PUCCH, PDCCH, etc.) and information elements may be identified by any suitable names, and therefore the various names assigned to these various channels and information elements are not intended to be limiting in any way.

[0075] In this disclosure, the terms "Mobile Station (MS)," "user terminal," "User Equipment (UE)," "terminal," and the like may be used interchangeably.

[0076] A mobile station may also be referred to by those skilled in the art as a subscriber station, mobile unit, subscriber unit, wireless unit, remote unit, mobile device, wireless device, wireless communication device, remote device, mobile subscriber station, access terminal, mobile terminal, wireless terminal, remote terminal, handset, user agent, mobile client, client, or some other suitable terminology.

[0077] As used in this disclosure, the terms "determining" and "determining" may encompass a wide variety of actions. "Determining" and "determining" may include, for example, judging, calculating, computing, processing, deriving, investigating, looking up, searching, inquiring (e.g., searching in a table, database, or other data structure), ascertaining, and the like. "Determining" and "determining" may also include receiving (e.g., receiving information), transmitting (e.g., sending information), input, output, accessing (e.g., accessing data in memory), and the like. Furthermore, "judgment" and "decision" can include regarding resolving, selecting, choosing, establishing, comparing, etc. as having been "judged" or "decided." In other words, "judgment" and "decision" can include regarding some action as having been "judged" or "decided." Furthermore, "judgment (decision)" can be interpreted as "assuming," "expecting," "considering," etc.

[0078] The terms "connected," "coupled," or any variation thereof, refer to any direct or indirect connection or coupling between two or more elements, and may include the presence of one or more intermediate elements between two elements that are "connected" or "coupled" to each other. The coupling or connection between elements may be physical, logical, or a combination thereof. For example, "connected" may be read as "access." As used in this disclosure, two elements may be considered to be "connected" or "coupled" to each other using one or more wires, cables, and / or printed electrical connections, as well as electromagnetic energy having wavelengths in the radio frequency range, microwave range, and optical (both visible and invisible) range, as some non-limiting and non-exhaustive examples.

[0079] As used in this disclosure, the phrase "based on" does not mean "based only on," unless expressly stated otherwise. In other words, the phrase "based on" means both "based only on" and "based at least on."

[0080] As used in this disclosure, any reference to an element using a designation such as "first," "second," etc. does not generally limit the quantity or order of those elements. These designations may be used in this disclosure as a convenient method of distinguishing between two or more elements. Thus, a reference to a first and a second element does not imply that only two elements may be employed or that the first element must in some way precede the second element.

[0081] When the terms "include," "including," and variations thereof are used in this disclosure, these terms are intended to be inclusive, similar to the term "comprising." Furthermore, when the term "or" is used in this disclosure, it is not intended to be an exclusive or.

[0082] In this disclosure, where articles are added by translation, such as a, an, and the in English, the disclosure may include that the nouns following these articles are in the plural form.

[0083] In the present disclosure, the term "A and B are different" may mean "A and B are different from each other." The term may also mean "A and B are each different from C." Terms such as "separate" and "coupled" may also be interpreted in the same way as "different."

[0084] 100...prompt generation device, 200...user terminal, 300...LLM, 400...ethics AI, 101...request acquisition unit, 102...prompt generation unit, 103...answer acquisition unit, 104...ethics check unit, 105...answer output unit.

Claims

1. An apparatus comprising: a prompt generation unit that generates a prompt to instruct a generation AI to generate suggested content in accordance with input from a user, based on the input, and sends the prompt to the generation AI; an answer acquisition unit that acquires an answer from the generation AI; and a content check unit that acquires check results for the answer, wherein the prompt generation unit requests a new answer from the generation AI based on the check results.

2. The device according to claim 1, wherein the content check unit checks whether the answer contains inappropriate language, inappropriate content, or is ethically questionable.

3. The device according to claim 1, wherein the content check unit checks whether the response contains personal information.

4. The device according to claim 1, wherein the prompt generator generates a new prompt based on the answer and requests the new answer.

5. The device according to claim 4, wherein the prompt generation unit generates a new prompt indicating that the answer is NG if the answer is NG.

6. The device according to claim 1, further comprising: a response output unit that outputs the response to the user when the check result is determined to be normal.

7. The device according to claim 1, wherein the checking unit transmits the answer to a checking AI and obtains the check result.

8. A method comprising: a prompt generation step of generating a prompt to instruct a generation AI to perform ethical judgment processing based on input from a user, and sending the prompt to the generation AI; an answer acquisition step of acquiring an answer from the generation AI; and an ethical check step of acquiring the results of an ethical check on the answer, wherein the prompt generation step again requests a new answer from the generation AI based on the check results.