Information processing apparatus, information processing method, and information processing program

Abstract

To provide an information processing apparatus, an information processing method, and an information processing program capable of supporting information generation using an appropriate prompt in accordance with a request from a user.SOLUTION: An information processing apparatus includes a selection unit, a generation unit, and a provision unit. The selection unit selects a prompt according toa request from a user, from among a plurality of prompts for which details are concealed from a user. The generation unit inputs, to a generative AI, information including the prompt selected by the selection unit, to cause the generative AI to generate information tailored to the prompt selected by the selection unit. The provision unit provides the information generated by the generation unit or information based on the information, as provision information, to the user.SELECTED DRAWING: Figure 3

Description

Technical Field

[0001] The present invention relates to an information processing apparatus, an information processing method, and an information processing program.

Background Art

[0002] In recent years, techniques for generating information using generative AI (Artificial Intelligence) have been known. For example, Patent Document 1 discloses a technique using large language models (LLMs) that generates a prompt by adding a valid sentence as an additional sentence to an input question sentence within a determined character limit. A prompt is information input to generative AI and indicates, for example, instructions or requests given to generative AI to perform a specific task.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] However, in the above prior art, since additional sentences are automatically added, there is room for further improvement in assisting the use of more appropriate prompts according to the user's requests.

[0005] The present application has been made in view of the above, and an object thereof is to provide an information processing apparatus, an information processing method, and an information processing program that can assist in using more appropriate prompts according to the user's requests.

Means for Solving the Problems

[0006] The information processing apparatus according to the present application includes a selection unit, a generation unit, and a provision unit. The selection unit selects a prompt corresponding to a request from a user from among a plurality of prompts whose details are concealed from the user. The generation unit inputs information including the prompt selected by the selection unit to a generation AI, and causes the generation AI to generate information corresponding to the prompt selected by the selection unit. The provision unit provides the user with the information generated by the generation unit or information based on the information as provision information.

Effect of the Invention

[0007] According to one aspect of the embodiment, there is an effect that it is possible to assist in using a more appropriate prompt according to the user's request.

Brief Description of the Drawings

[0008]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Mode for Carrying Out the Invention

[0009] Hereinafter, embodiments for implementing an information processing apparatus, an information processing method, and an information processing program according to the present application (hereinafter referred to as "embodiments") will be described in detail with reference to the drawings. Note that the information processing apparatus, the information processing method, and the information processing program according to the present application are not limited by these embodiments. Also, in the following embodiments, the same parts are denoted by the same reference numerals, and redundant descriptions are omitted.

[0010] 〔1. An example of information processing〕 First, an example of information processing according to the embodiment will be described with reference to FIG. 1. FIG. 1 is a diagram for explaining the information processing according to the embodiment.

[0011] The information processing apparatus 1 shown in FIG. 1 is an information processing apparatus that cooperates with each terminal device 2 of the user U and provides various types of information to the user U online, and is realized by, for example, one or more servers or a cloud system. The terminal device 2 is, for example, a smartphone, a tablet, or a personal computer.

[0012] As shown in FIG. 1, the information processing apparatus 1 receives a search request from the user U (step S1). For example, the information processing apparatus 1 receives a search request from the user U by receiving a search request transmitted from the terminal device 2 of the user U.

[0013] The search request includes user-related information including user setting information, which is information set by the user U, and user information, which is information of the user U who is the user of the confidentiality prompt described later. The user U who is the user of the confidentiality prompt is not limited to the user U of the terminal device 2 that transmitted the search request, and may be a user U other than the user U of the terminal device 2 that transmitted the search request. Thereby, the user U of the terminal device 2 can provide a confidentiality prompt for another user U.

[0014] The user setting information is, for example, information such as the string "prompt that can ~~~" or the string "prompt to ~~~". "~~~" is a string indicating the operations, processes, etc. that the user U desires from the generative AI (Artificial Intelligence).

[0015] For example, the user setting information is information such as the string "prompt that can summarize the latest information of OOO", the string "prompt that can proofread text", or the string "prompt that functions as a shopping assistant". "OOO" is a specific target (e.g., a person). Note that the user setting information is not limited to the above examples. For example, the user setting information may be information of a string that more specifically indicates the content of operations, processes, etc. desired by the user U.

[0016] The user information includes the identification information of the user U. The identification information of the user U is, for example, the ID (Identifier) of the user U or the cookie of the browser of the terminal device 2, but is not limited to such examples.

[0017] The generative AI is, for example, a text generation AI, an image generation AI, or a multimodal generation AI. The text generation AI is, for example, a large language model trained to estimate and output the next token from the input token sequence, and is, for example, a transformer-based model, an RNN (Recurrent Neural Network)-based model, etc., but may also be a hybrid model thereof. Further, the text generation AI may be a composite system combined with an identification mechanism for preventing unauthorized use.

[0018] Transformer-based models include, for example, but are not limited to, GPT (Generative Pre-trained Transformer) (registered trademark) and PaLM2 (Pathways Language Model Version 2). RNN-based models include, for example, but are not limited to, RWKV (Receptance Weighted Key Value).

[0019] Image generation AI is AI that generates images from text, and includes, for example, but is not limited to, StackGAN (Generative Adversarial Networks), AttnGAN, T2I (Text-to-Image) with Transformers, Diffusion models, etc. Examples of Diffusion models include DALL-E and Stable-Diffusion.

[0020] Multimodal generation AI is, for example, generative AI that generates at least one of text, images, and audio from at least one of text, images, and audio. Multimodal generation AI includes, for example, but is not limited to, GPT-4 Turbo with vision, gemini, CM3Leon (Chameleon Multimodal Model), etc.

[0021] It is desirable that the generative AI be trained so as not to include personal information, etc. in its generation results. The generative AI is arranged in an external information processing device, and the information processing device 1 uses the generative AI via an API (Application Programming Interface), but the generative AI may also be arranged within the information processing device 1.

[0022] Next, the information processing apparatus 1 searches for two or more confidential prompts that respond to the search request from among a plurality of confidential prompts (step S2). Each confidential prompt is a prompt whose details are concealed from the user U, and the user U cannot grasp the details of the confidential prompt. Each confidential prompt is, for example, a prompt provided by a prompt provider who is a business operator or operator different from the provider of the above-described generative AI. Note that the confidential prompt may be a prompt provided by the user U as a prompt provider.

[0023] The confidential prompt includes, for example, but is not limited to, some or all of the string information indicating the definition of the task executed by the generative AI, the constraint conditions of the task in the generative AI, the definition of the behavior and tone of the generative AI, the output format of the generative AI, etc.

[0024] For example, when the generative AI is GPT of OpenAI (registered trademark), the confidential prompt includes the string information included in the input information input to the generative AI in association with the information indicating the role. The role is, for example, a system, a user, etc. The confidential prompt is a prompt including at least one of a system message with the role being the system and a user message with the role being the user.

[0025] The system message is, for example, a prompt whose role indicates the role of the generative AI, and the user message is, for example, a prompt whose role indicates the message of the user. The system message includes, for example, but is not limited to, the string information indicating the definition of the task executed by the generative AI, the constraint conditions of the task in the generative AI, the definition of the behavior and tone of the generative AI, the output format of the generative AI, the operation example of the generative AI, etc., and may include a part of these information.

[0026] In addition, the system message may be, for example, a message indicating a scenario, and in this case, it may further include information on a character string indicating the end condition of a task in the above-described generative AI, for example.

[0027] The user message may include information on a character string indicating the definition of a task executed by the generative AI, the constraint conditions of the task in the generative AI, the definition of the behavior and tone of the generative AI, the output format of the generative AI, an operation example of the generative AI, etc., or may include a part of these information.

[0028] In addition, when the generative AI is GPT of OpenAI, the secret prompt may further include an assistant message whose role is an assistant. The assistant message is an output message by the generative AI or a message treated as an output message.

[0029] In step S2, the information processing apparatus 1 has a secret prompt database including a plurality of secret prompts, and identifies a secret prompt corresponding to a search request by searching for and acquiring a secret prompt corresponding to the search request from such a secret prompt database.

[0030] The secret prompt database includes, for each secret prompt, information on the secret prompt, including, for example, identification information of the secret prompt, the title of the secret prompt, the heading information of the secret prompt, the vector information of the heading information of the secret prompt, information indicating the secret prompt, etc. Such a secret prompt database includes a searchable secret prompt corresponding to a search query.

[0031] The title information of the confidential prompt is information indicating a description text (such as an overview of the function, etc.) that explains the function of the confidential prompt. The vectorization of the title information of the confidential prompt is performed, for example, by embedding using a sentence embedding model (such as a transformer-based model). The vector information of the title information of the confidential prompt is represented by a vector of, for example, several hundred dimensions, but is not limited to such an example.

[0032] The embedding by the sentence embedding model is, for example, an embedding by text-embedding-ada provided by OpenAI, BERT (Bidirectional Encoder Representations from Transformers), etc., but is not limited to such an example.

[0033] Note that the vectorization of the title information of the confidential prompt is not limited to the embedding by the sentence embedding model. For example, the vectorization of the confidential prompt may be performed by Doc2Vec, the average of word embeddings, etc. For word embeddings, for example, Word2Vec, fastText, etc. are used.

[0034] The information processing device 1 vectorizes the user setting information included in the search request in the same way as the vectorization of the title information of the confidential prompt. Then, the information processing device 1 acquires, as search results, the confidential prompts corresponding to the top m (m is 2 or more) title information in descending order of the vector similarity with the user setting information being equal to or higher than a threshold value or having a high similarity from the confidential prompt database.

[0035] Note that the confidential prompt database may include, instead of the vector information of the title information of the confidential prompt, information on each of the confidential prompt or a plurality of keywords included in its title information. In this case, the information processing device 1 acquires, as search results, the confidential prompts corresponding to the confidential prompt or title information including one or more terms included in the user setting information included in the search request from the confidential prompt database.

[0036] In addition, the prompt database may include the vector information of the prompt instead of or in addition to the vector information of the heading information of the prompt. In this case, the information processing apparatus 1 acquires, from the prompt database, the top m (m is 2 or more) prompts as prompt candidates in descending order of the vector similarity with the user setting information being equal to or higher than the threshold value or having a high similarity.

[0037] In addition, the confidential prompt database may include information indicating the charging amount of the confidential prompt, information indicating the number of selections of the confidential prompt, information indicating the evaluation of the confidential prompt, and the like. The charging amount of the confidential prompt is the fee that the user U pays for using the confidential prompt.

[0038] Subsequently, the information processing apparatus 1 provides the user U with the information of two or more confidential prompts obtained in the search in step S2 by transmitting the information of two or more confidential prompts obtained in the search in step S2 to the terminal device 2 (step S3).

[0039] The terminal device 2 displays the information of two or more confidential prompts transmitted from the information processing apparatus 1. The information of the confidential prompt displayed on the terminal device 2 includes the title of the confidential prompt and the heading information of the confidential prompt. The user U confirms the titles and heading information of two or more confidential prompts and selects one or more confidential prompts from the two or more confidential prompts displayed on the terminal device 2 by operating the terminal device 2.

[0040] In addition, the information of the confidential prompt displayed on the terminal device 2 may include information indicating the charging amount of the confidential prompt, information indicating the number of selections of the confidential prompt, information indicating the evaluation of the confidential prompt, and the like. In this case, the user U can confirm information indicating the charging amount of the confidential prompt, information indicating the number of selections of the confidential prompt, information indicating the evaluation of the confidential prompt, and the like, and select one or more confidential prompts from the two or more confidential prompts displayed on the terminal device 2.

[0041] When the user U designates one or more of the two or more confidential prompts retrieved by the information processing apparatus 1 on the terminal device 2, the terminal device 2 transmits a selection request including information for specifying the one or more confidential prompts designated by the user U to the information processing apparatus 1 (step S4).

[0042] Also, when additional information is required as additional information for using the confidential prompt designated by the user U, the user U inputs the additional information by operating the terminal device 2. The additional information is an additional prompt when the generation AI is a text generation AI, and is at least one of an additional prompt and additional image data when the generation AI is a multimodal generation AI.

[0043] For example, when the confidential prompt designated by the user U is a prompt that requires additional information, the user U can input the additional information into the terminal device 2 by operating the terminal device 2. Thereby, a selection request including information for specifying the one or more confidential prompts designated by the user U and the additional information is transmitted from the terminal device 2 to the information processing apparatus 1.

[0044] The prompt that requires additional information is, for example, information in the character string "You are an experienced travel planner. For a specific request from the user, propose a 3-day travel plan. The plan includes places to visit, accommodation facilities, dining places, and activities." In this case, the additional information is, for example, information in the character string "Please create a travel plan for Okinawa. I am interested in history and gourmet. The budget is 200,000 yen per person."

[0045] Also, the prompt that requires additional information is, for example, information in the character string "You are an excellent corrector. Detect up to 20 misspellings from the input text and propose a correction. The output format is the incorrect text, the correct text, and an explanation of the error." In this case, the additional information is information on the character string to be corrected.

[0046] When the information processing device 1 receives a selection request transmitted from the terminal device 2, it selects one or more anonymization prompts specified by the selection request from among a plurality of anonymization prompts (step S5). The one or more anonymization prompts selected by the information processing device 1 are associated with the user U of the terminal device 2 that transmitted the selection request.

[0047] Subsequently, the information processing device 1 inputs information including the selection prompt, which is the anonymization prompt selected in step S5, to the generation AI, and causes the generation AI to generate generated information that is information corresponding to the selection prompt (step S6).

[0048] For example, assume that the selection prompt is a prompt that does not require additional information. In this case, the information processing device 1 inputs information including the selection prompt to the generation AI, and causes the generation AI to generate generated information that is information corresponding to the selection prompt.

[0049] A selection prompt that does not require additional information is, for example, a prompt that collects information on a specific target whose state changes (e.g., a specific disaster, a specific person, a specific object) and instructs the generation of current information or summary information on the specific target.

[0050] In this case, the information processing device 1 collects information on a specific target whose state changes (e.g., a specific disaster, a specific person, a specific object) according to the selection prompt, for example, and inputs, as input information, information including the collected information and instruction information indicating an instruction to execute a specific process on the information, and causes the generation AI to generate generated information that is information corresponding to the selection prompt. The specific process is, for example, a process of summarizing the collected information in a specific format, a process of summarizing the collected information, etc., but is not limited to such examples.

[0051] A selection prompt that does not require additional information is, for example, a prompt including information on a character string that instructs a specific process of repeatedly outputting a character string according to a predetermined condition or repeatedly outputting an image according to a predetermined condition.

[0052] Also, assume that the selection prompt is a prompt that requires additional information. In this case, the information processing apparatus 1 inputs information including the information containing the selection prompt and the additional information included in the selection request into the generation AI, and causes the generation AI to generate generated information that is information corresponding to the selection prompt and the additional information.

[0053] The selection prompt that requires additional information is, for example, a prompt that instructs the execution of a specific task with respect to the additional information. The specific task is, for example, but not limited to, summarization of additional information, analysis of additional information (e.g., sentiment analysis, etc.), translation of additional information, improvement or proofreading of additional information, generation of an article or a story regarding the additional information, and the like.

[0054] The selection prompt that requires additional information may be, for example, a prompt for executing a specific task in a chat format with the user U. In this case, the specific task is a task of conducting Q&A (Question and Answer) regarding a specific category (e.g., economy, education, etc.) in a chat format, and causing the user U to input information indicating a question sentence as additional information, and causing the generation AI to generate information indicating an answer sentence to the question sentence.

[0055] Also, the specific task may be a task of sequentially causing the user U to input, as additional information, information necessary for the user U to create information on a specific target (e.g., various application documents, novels, cooking recipes, blogs, etc.), and causing the generation AI to generate information on the specific target based on these additional information.

[0056] The anonymized prompt selected as the selection prompt may be a prompt that includes intention definition information for extracting the intention type and intention content. The intention definition information includes, for example, instruction information including an instruction to extract the intention type and intention content from the user input information, and definition information including the intention type and information defining the intention type. When the information processing apparatus 1 uses, for example, an API provided by OpenAI, it can cause the generative AI to generate intention information by using the function calling function.

[0057] The intention type is, for example, the type of intention for additional information, and the information indicating the intention type is, for example, information specifying a function or function corresponding to the intention type. The intention content is the content of the intention for additional information, and the information indicating the intention content is, for example, information indicating an argument of a function or a parameter of a function corresponding to the intention type.

[0058] In this case, intention information including information indicating the intention type and information indicating the intention content is generated by the generative AI. The information processing apparatus 1 uses the intention information generated by the generative AI to obtain information from an external information processing apparatus or an internal storage unit, etc., and inputs information including the obtained information and instruction information indicating an instruction to generate generated information using the information to the generative AI, and can cause the generative AI to generate generated information.

[0059] Further, when the selection prompt included in the information processing apparatus 1 includes a message indicating a scenario, until the end condition of the task is satisfied, the process of inputting information to the generative AI according to the scenario and obtaining an output is repeated, and based on such a processing result, generated information can also be generated. In this case, the information processing apparatus 1 includes, in the input information input to the generative AI, an input / output history including information that was input to the generative AI in the past and information that was output by the generative AI in the past when the task was executed.

[0060] For example, until the end condition of the task is satisfied, the information processing apparatus 1 can cause the generative AI to generate generated information by inputting information including the input / output history into the generative AI while changing the content to be generated by the generative AI according to the scenario.

[0061] The information input into the generative AI according to the scenario is information collected from an external information processing apparatus or an internal storage unit, or additional information input by the user U. For example, until the end condition of the task is satisfied, the information processing apparatus 1 collects information from an external information processing apparatus or an internal storage unit according to the scenario, and causes the generative AI to generate generated information based on the collected information.

[0062] Also, until the end condition of the task is satisfied, the information processing apparatus 1 can cause the user U to input additional information according to the scenario, and cause the generative AI to generate generated information based on this additional information.

[0063] When the information processing apparatus 1 selects a plurality of anonymized prompts in step S5, the information processing apparatus 1 causes the generative AI to generate generated information for each selected anonymized prompt. Thereby, the information processing apparatus 1 can obtain generated information for each selected prompt.

[0064] Subsequently, the information processing apparatus 1 provides the user U with the generated information generated in step S6 or information based on the generated information as provided information (step S7). The information processing apparatus 1 transmits the generated information generated in step S6 or information based on the generated information to the terminal device 2 as provided information, thereby providing the user U with the generated information generated in step S6 or information based on the generated information as provided information. Thereby, it is possible to provide the user U with the generated information generated using the anonymized prompt or information based on the generated information while keeping the detailed content of the anonymized prompt confidential.

[0065] When the information generated by the generative AI as the generated information is the intended information, the information processing apparatus 1 can acquire information from an external information processing apparatus or an internal storage unit or the like using such intended information, and generate provided information based on the acquired information.

[0066] Further, when the information processing apparatus 1 causes the generative AI to generate the generated information for each selection prompt, the information processing apparatus 1 provides the provided information for each selection prompt to the user U. In this case, the terminal device 2 displays the provided information for each selection prompt, and the user U can compare the provided information for each selection prompt.

[0067] Subsequently, the information processing apparatus 1 receives information indicating the evaluation of the user U for the provided information (step S8). The user U can confirm the provided information displayed on the terminal device 2, operate the terminal device 2, and input information indicating the evaluation for the provided information.

[0068] The evaluation of the provided information is, for example, a 10 - level evaluation, but it may be an evaluation of 9 levels or less, or an evaluation of 11 levels or more. Also, the information indicating the evaluation may be information in the form of a character string such as a review or a comment.

[0069] Subsequently, the information processing apparatus 1 performs a charging process for the user U who provided the provided information in step S7 (step S9). The charging process is, for example, a process of payment by the user U of the charging amount associated with the confidential prompt used for generating the provided information provided to the user U in step S7.

[0070] The information processing apparatus 1 can perform the charging process, for example, by transmitting charging information including information indicating the charging amount and information identifying the user U to the apparatus of a financial institution, or by transmitting charging information including information indicating the charging amount and information identifying the credit card number of the user U to the apparatus of a credit card company.

[0071] In addition, the user U can pay the consideration for the confidential prompt in points instead of or in addition to a specific currency. For example, the information processing device 1 can also perform a charging process by transmitting charging information including information indicating the number of points corresponding to the charge amount and information identifying the user U to the device of the point management company.

[0072] Note that the points are, for example, rewards given to the user U by a point management company or the like when the user U shops with a credit card or cash, or when the user U subscribes to a specific service, but are not limited to such examples.

[0073] Subsequently, the information processing device 1 determines the charge amount for the use of the confidential prompt (step S10). For example, the information processing device 1 determines the charge amount based on the evaluation of the provided information by each user U. The information processing device 1 determines a higher charge amount for a confidential prompt with a higher average value of the evaluation of the provided information by each user U.

[0074] The information processing device 1 can determine the charge amount based on the evaluation of the provided information by each user U, with reference to the reference amount set by the provider of the confidential prompt. For example, the information processing device 1 determines a lower charge amount than the reference amount for a confidential prompt with an average value of the evaluation of the provided information by each user U lower than the reference value. The reference value is 5 when the evaluation is on a 10-point scale.

[0075] In addition, the information processing device 1 can also determine the charge amount based on the number of times the user U uses the confidential prompt. For example, the information processing device 1 determines a higher charge amount for a confidential prompt with a larger number of times the user U uses the confidential prompt.

[0076] In addition, the information processing device 1 can also determine the charging amount based on the evaluation of the provided information by each user U and the number of times the user U uses the anonymization prompt. For example, the information processing device 1 determines a higher charging amount for anonymization prompts with a larger score obtained by weighted addition of the average value of the evaluation of the provided information by each user U and the number of times the user U uses the anonymization prompt.

[0077] In addition, the information processing device 1 can also receive a provisional selection request transmitted from the terminal device 2. The provisional selection request includes information indicating two or more anonymization prompts specified by the user U as selection candidates. When the information processing device 1 receives the provisional selection request, it selects two or more anonymization prompts indicated by the provisional selection request as two or more selection prompt candidates.

[0078] Also, when additional information is required as additional information for using the selection prompt candidates, the provisional selection request includes the additional information for each selection prompt candidate in addition to the specific information that is information indicating two or more selection prompt candidates.

[0079] Then, the information processing device 1 inputs the information including the selection prompt candidates to the generation AI, and causes the generation AI to generate candidate generation information that is information corresponding to the selection prompt candidates for each selection prompt candidate. The method for generating the candidate generation information is the same as the method for generating the generation information described above.

[0080] The information processing device 1 provides the selection information to the user U by transmitting the generated candidate generation information or the selection information including the information based on each candidate generation information to the terminal device 2. In the selection information, the candidate generation information or the information based on the candidate generation information is included for each selection prompt candidate in association with the information of the selection prompt candidate (for example, information indicating the title of the selection prompt candidate or heading information). The candidate generation information and the information based on the candidate generation information are the same information as the provided information described above, but may be a part of the provided information, and may be described as candidate provided information below.

[0081] When the terminal device 2 receives the selection information transmitted from the information processing device 1, it displays such selection information. As a result, the user U can compare the candidate-providing information between two or more selection prompt candidates by checking the candidate-providing information generated using each of the two or more selection prompt candidates, and can specify and use a more appropriate concealment prompt as the selection prompt.

[0082] The user U checks the candidate-providing information corresponding to each selection prompt candidate included in the selection information displayed on the terminal device 2, and designates one or more selection prompt candidates by operating the terminal device 2. In this case, the terminal device 2 transmits a selection request including information identifying one or more selection prompt candidates designated by the user U to the information processing device 1 in the same manner as the processing in step S4.

[0083] In this way, the information processing device 1 selects a prompt corresponding to the request from the user U from among a plurality of prompts whose details are concealed from the user U, inputs information including the selected prompt, which is the selection prompt, to the generation AI, and causes the generation AI to generate information corresponding to the selection prompt. Then, the information processing device 1 provides the user U with the information generated by the generation AI or information based on the said information as the provided information. Thereby, the information processing device 1 can perform information generation using a more appropriate prompt corresponding to the request from the user U.

[0084] Hereinafter, the configuration of the information processing system including the information processing device 1 and the terminal device 2 that perform such processing will be described in detail.

[0085] 〔2. Configuration of Information Processing System〕 FIG. 2 is a diagram showing an example of the configuration of the information processing system according to the embodiment. As shown in FIG. 2, the information processing system 100 according to the embodiment includes an information processing device 1, a plurality of terminal devices 2, and a terminal device 3.

[0086] The plurality of terminal devices 2 are used by different users U. The terminal device 3 is, for example, the terminal device of the operator O that provides the anonymized prompt. In the example shown in FIG. 2, one terminal device 3 is shown, but when there are a plurality of operators O that provide the anonymized prompt, the information processing system 100 includes a plurality of terminal devices 3.

[0087] The terminal devices 2 and 3 are, for example, notebook PCs (Personal Computers), desktop PCs, smartphones, tablet PCs, or wearable devices. The wearable device is, for example, smart glasses or a smartwatch, but is not limited to such examples.

[0088] Each of the information processing device 1, the terminal device 2, and the terminal device 3 is connected to be communicable with each other by wire or wirelessly via the network N. Note that the information processing system 100 shown in FIG. 2 may include a plurality of information processing devices 1 and the like.

[0089] The network N includes, for example, a WAN (Wide Area Network) such as the Internet and a mobile communication network such as LTE (Long Term Evolution), 4G (4th Generation), and 5G (5th Generation: the 5th generation mobile communication system).

[0090] The terminal devices 2 and 3 can be connected to the network N via a mobile communication network, short-range wireless communication such as Bluetooth (registered trademark), or wireless LAN (Local Area Network), and communicate with the information processing device 1 and the like.

[0091] 〔3. Configuration of Information Processing Device 1〕 FIG. 3 is a diagram showing an example of the configuration of the information processing device 1 according to the embodiment. As shown in FIG. 3, the information processing device 1 includes a communication unit 10, a storage unit 11, and a processing unit 12.

[0092] 〔3.1. Communication Unit 10〕 The communication unit 10 is realized by, for example, a communication module or a NIC (Network Interface Card). The communication unit 10 is connected to the network N either wired or wirelessly and transmits and receives information to and from various other devices. For example, the communication unit 10 transmits and receives information to and from each of the terminal devices 2 and 3 via the network N.

[0093] [3.2. Memory unit 11] The memory unit 11 is realized by, for example, semiconductor memory elements such as RAM (Random Access Memory) and flash memory, or storage devices such as hard disks and optical disks. The memory unit 11 has a user information storage unit 20 and a confidential prompt storage unit 21.

[0094] [3.2.1. User information storage unit 20] The user information storage unit 20 stores user information including information about the user U. FIG. 4 is a diagram showing an example of a user information table stored in the user information storage unit 20 of the information processing apparatus 1 according to the embodiment. As shown in FIG. 4, the user information table stored in the user information storage unit 20 includes items such as "user ID", "attribute information", and "behavior history".

[0095] The "user ID" is identification information for identifying the user U. The "attribute information" is attribute information of the user U corresponding to the "user ID", and includes, for example, information on psychographic attributes and information on demographic attributes. Demographic attributes are, for example, gender, age, place of residence, and occupation, and psychographic attributes are objects of interest such as travel, clothing, cars, and religion, lifestyle, thoughts, and trends of thoughts.

[0096] The "action history" is the action history of user U corresponding to the "user ID". The action history of user U includes, for example, the identification information of the anonymized prompt selected by the processing unit 12 for user U, the identification information of the anonymized prompt purchased by user U, the identification information of the anonymized prompt used by user U, information indicating the number of times each anonymized prompt used by user U, etc.

[0097] [3.2.2. Anonymized Prompt Storage Unit 21] The anonymized prompt storage unit 21 stores various types of information related to anonymized prompts. FIG. 5 is a diagram showing an example of an anonymized prompt table stored in the anonymized prompt storage unit 21 of the information processing apparatus 1 according to the embodiment.

[0098] In the example shown in FIG. 5, the anonymized prompt table stored in the anonymized prompt storage unit 21 is an example of the above-described anonymized prompt database, and includes information on items such as "prompt ID", "anonymized prompt", "related information", "selection count", "usage count", and "evaluation information". The "prompt ID" is an identifier for identifying an anonymized prompt, and is information assigned to each anonymized prompt.

[0099] The "anonymized prompt" is the anonymized prompt associated with the "prompt ID". The "related information" is information related to the anonymized prompt associated with the "prompt ID", and includes, for example, the identification information of the anonymized prompt, the title of the anonymized prompt, the heading information of the anonymized prompt, the vector information of the heading information of the anonymized prompt, the vector information of the anonymized prompt, information indicating the anonymized prompt, information indicating the charge amount of the anonymized prompt, etc. The charge amount of the anonymized prompt is the fee that user U pays for the use of the anonymized prompt.

[0100] The "selection count" is information indicating the number of times the anonymized prompt associated with the "prompt ID" has been selected and associated with user U. The "usage count" is the number of times the anonymized prompt associated with the "prompt ID" has been used by user U.

[0101] "Evaluation information" is information indicating the evaluation of each user U for the provided information generated by the processing unit 12 using the confidential prompt associated with the "prompt ID". The evaluation of the provided information is, for example, a 10-point evaluation, but it may also be an evaluation of 9 points or less, or an evaluation of 11 points or more. Further, the information indicating the evaluation may be information in the form of a character string such as a review or a comment.

[0102] Although not shown in the figure, the confidential prompt table also includes information indicating the number of purchases, which is the number of times the confidential prompt associated with the "prompt ID" has been purchased by the user U.

[0103] 〔3.3. Processing Unit 12〕 The processing unit 12 is a controller and is realized, for example, by a processor such as a CPU (Central Processing Unit) or an MPU (Micro Processing Unit), when various programs (corresponding to an example of an information processing program) stored in the storage device inside the information processing apparatus 1 are executed with a RAM or the like as a work area.

[0104] Further, the processing unit 12 is a controller and may also be realized by an integrated circuit such as an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or a GPGPU (General Purpose Graphic Processing Unit).

[0105] As shown in FIG. 3, the processing unit 12 includes a reception unit 30, an acquisition unit 31, a search unit 32, a selection unit 33, a generation unit 34, a provision unit 35, a charging unit 36, and a determination unit 37, and realizes or executes the functions and operations of information processing described below. Note that the internal configuration of the processing unit 12 is not limited to the configuration shown in FIG. 3, and any other configuration may be used as long as it can perform the information processing described later.

[0106] [[3.3.1. Reception Unit 30]] The reception unit 30 receives various information and requests. For example, the reception unit 30 receives a prompt list request from the user U. The reception unit 30 receives a prompt list request from the user U by receiving, for example, a prompt list request transmitted from the terminal device 2 of the user U.

[0107] Also, the reception unit 30 receives a search request from the user U. The reception unit 30 receives a search request from the user U by receiving, for example, a search request transmitted from the terminal device 2 of the user U.

[0108] The search request includes user-related information including user setting information which is information set by the user U and user information which is information of the user U who is the user of the confidential prompt. The user U who is the user of the confidential prompt is not limited to the user U of the terminal device 2 that transmitted the search request, and may be a user U other than the user U of the terminal device 2 that transmitted the search request. Thereby, the user U of the terminal device 2 can provide a confidential prompt for another user U.

[0109] The user setting information is, for example, information of the character string "prompt that can ~~~" or information of the character string "prompt that ~~~". "~~~" is a character string indicating an operation or process that the user U desires for the generative AI.

[0110] For example, the user setting information is information of the character string "prompt that summarizes the latest information of 〇〇〇", information of the character string "prompt that can proofread an article", or information of the character string "prompt that functions as a shopping assistant". "〇〇〇" is a specific target (for example, a person). Note that the user setting information is not limited to the above examples. For example, the user setting information may be information of a character string that more specifically indicates the content of an operation or process desired by the user U.

[0111] The user information includes the identification information of user U. The identification information of user U is, for example, the ID of user U or the cookie of the browser of the terminal device 2, etc., but is not limited to such examples.

[0112] The generative AI is, for example, text generative AI, image generative AI, or multimodal generative AI. The text generative AI is, for example, a large language model trained to estimate and output the next token from the input token sequence, and is, for example, a model based on transformers or an RNN-based model, etc., but may also be a hybrid model thereof. Further, the text generative AI may be a composite system combined with an identification mechanism for preventing unauthorized use.

[0113] The model based on transformers is, for example, GPT, BARD, etc., but is not limited to such examples. The RNN-based model is, for example, RWKV, etc., but is not limited to such examples.

[0114] The image generative AI is an AI that generates an image from text, and is, for example, StackGAN, AttnGAN, T2I with Transformers, Diffusion model, etc., but is not limited to such examples. Examples of the Diffusion model include DALL-E and Stable-Diffusion.

[0115] The multimodal generative AI is, for example, a generative AI that generates at least one of text, image, and voice from at least one of text, image, and voice. The multimodal generative AI is, for example, GPT-4V, CM3Leon, etc., but is not limited to such examples.

[0116] Note that it is desirable for the generative AI to be trained not to include personal information, etc. in its generation results. The generative AI is arranged in an external information processing device, and the information processing device 1 uses the generative AI via an API, but the generative AI may also be arranged within the information processing device 1.

[0117] In addition, the reception unit 30 receives a selection request from the user U. The reception unit 30 receives the selection request from the user U, for example, by receiving a selection request transmitted from the terminal device 2 of the user U.

[0118] The selection request includes, for example, specific information that is information for specifying one or more confidential prompts specified by the user U from among two or more confidential prompts retrieved by the search unit 32. Also, when additional information is required to use the confidential prompt specified by the user U, the selection request includes the additional information.

[0119] The additional information is an additional prompt when the generation AI is a text generation AI, and is at least one of an additional prompt and additional image data when the generation AI is a multimodal generation AI.

[0120] For example, when the confidential prompt specified by the user U is a prompt that requires additional information, the user U can input the additional information into the terminal device 2 by operating the terminal device 2. Thereby, a selection request including information for specifying one or more confidential prompts specified by the user U and the additional information is transmitted from the terminal device 2 to the information processing device 1.

[0121] In addition, the reception unit 30 receives a provisional selection request from the user U. The reception unit 30 receives the provisional selection request from the user U, for example, by receiving a provisional selection request transmitted from the terminal device 2 of the user U. The provisional selection request includes information indicating two or more selection prompt candidates that are two or more confidential prompts specified by the user U as selection candidates.

[0122] Also, when additional information, which is additional information for using the selection prompt candidates, is required, the provisional selection request includes the additional information for each of the two or more selection prompt candidates in addition to the specific information that is information indicating the two or more selection prompt candidates.

[0123] In addition, the reception unit 30 receives a purchase request from the user U. For example, the reception unit 30 receives a purchase request from the user U by receiving a purchase request transmitted from the terminal device 2 of the user U. The purchase request includes information indicating one or more confidential prompts selected by the user U as purchase targets.

[0124] When the reception unit 30 receives a purchase request from the user U, the reception unit 30 causes the user information storage unit 20 to store the identification information of the confidential prompt specified in the purchase request as the identification information of the purchase prompt, and updates the user information table.

[0125] In addition, the reception unit 30 receives a usage request from the user U. For example, the reception unit 30 receives a usage request from the user U by receiving a usage request transmitted from the terminal device 2 of the user U. For example, when the confidential prompt specified in the usage request from the user U is the purchase prompt of the user U, the reception unit 30 receives the usage request from the user U.

[0126] The usage request includes information indicating one or more confidential prompts selected by the user U as usage targets. In addition, when additional information, which is additional information for using the confidential prompt selected by the user U as a usage target, is required, the usage request includes the additional information for each confidential prompt in addition to the specific information, which is information indicating one or more confidential prompts.

[0127] In addition, the reception unit 30 receives evaluation information, which is information indicating the evaluation of the user U for the provided information. For example, the reception unit 30 receives the evaluation information by receiving the evaluation information transmitted from the terminal device 2 of the user U.

[0128] 〔3.3.2. Acquisition Unit 31〕 The acquisition unit 31 acquires various types of information from an external device or the like via the network N and the communication unit 10, or acquires various types of information from the storage unit 11.

[0129] When the acquisition unit 31 receives a search request from the reception unit 30, for example, it acquires the user-related information included in the search request received by the reception unit 30. The user-related information includes user setting information which is information set by the user U, and user information which is information of the user U who is the user of the anonymization prompt. Also, when the acquisition unit 31 receives a selection request from the reception unit 30, for example, it acquires the specific information and additional information included in the selection request received by the reception unit 30.

[0130] Also, when the acquisition unit 31 receives a provisional selection request from the reception unit 30, for example, it acquires the specific information and additional information included in the provisional selection request received by the reception unit 30.

[0131] Also, when the acquisition unit 31 receives a purchase request from the reception unit 30, for example, it acquires the specific information included in the purchase request received by the reception unit 30. Also, when the acquisition unit 31 receives a usage request from the reception unit 30, for example, it acquires the specific information and additional information included in the usage request received by the reception unit 30.

[0132] [3.3.3. Search unit 32] The search unit 32 searches for two or more anonymization prompts in response to the search request received by the reception unit 30. For example, the search unit 32 searches for two or more anonymization prompts using the explanatory text that explains the function of each anonymization prompt.

[0133] Each anonymization prompt is a prompt whose details are hidden from the user U, and the user U cannot grasp the details of the anonymization prompt. Each anonymization prompt is, for example, a prompt provided by a prompt provider who is a different business operator or operator from the provider of the above-described generative AI (for example, the operator of the information processing apparatus 1). Note that the anonymization prompt may be a prompt provided by the user U as the prompt provider.

[0134] The confidential prompt includes, but is not limited to, some or all of the string information indicating, for example, the definition of the task executed by the generative AI, the constraints of the task in the generative AI, the definition of the behavior and tone of the generative AI, the output format of the generative AI, etc.

[0135] For example, when using the API provided by OpenAI, the confidential prompt includes the string information included in the input information input to the generative AI in association with the information indicating the role. The role can be a system, a user, etc. The confidential prompt is a prompt including at least one of the system message with the role being the system and the user message with the role being the user.

[0136] The system message is, for example, a prompt with the role indicating the role of the generative AI, and the user message is, for example, a prompt with the role indicating the message of the user. The system message includes, but is not limited to, the string information indicating, for example, the definition of the task executed by the generative AI, the constraints of the task in the generative AI, the definition of the behavior and tone of the generative AI, the output format of the generative AI, examples of the operations of the generative AI, etc., and may include some of this information.

[0137] Also, the system message may be, for example, a message indicating a scenario, and in this case, it may further include, for example, the string information indicating the end condition of the task in the above-mentioned generative AI.

[0138] The user message may include the string information indicating the definition of the task executed by the generative AI, the constraints of the task in the generative AI, the definition of the behavior and tone of the generative AI, the output format of the generative AI, examples of the operations of the generative AI, etc., and may include some of this information.

[0139] In addition, when using the API provided by OpenAI, the confidential prompt may further include an assistant message with the role being "assistant". The assistant message is an output message generated by the generative AI or a message treated as an output message.

[0140] The search unit 32 selects a confidential prompt corresponding to the search request by searching for and obtaining the confidential prompt corresponding to the search request from the confidential prompt table stored in the confidential prompt storage unit 21.

[0141] As described above, the confidential prompt table contains, for each confidential prompt, confidential prompt information including, for example, the identification information of the confidential prompt, the title of the confidential prompt, the heading information of the confidential prompt, the vector information of the heading information of the confidential prompt, and the information indicating the confidential prompt. The confidential prompt table contains searchable confidential prompts corresponding to the search query.

[0142] The heading information of the confidential prompt is information indicating a description text (e.g., an outline of the function, etc.) that explains the function of the confidential prompt. The vectorization of the heading information of the confidential prompt is performed, for example, by embedding (Embedding) using a text embedding model (e.g., a transformer-based model). The vector information of the heading information of the confidential prompt is represented by a vector of, for example, several hundred dimensions, but is not limited to such an example.

[0143] The embedding by the text embedding model is, for example, an embedding by text-embedding-ada, BERT, etc. provided by OpenAI, but is not limited to such an example.

[0144] Note that the vectorization of the heading information of the confidential prompt is not limited to embedding by a sentence embedding model. For example, the vectorization of the confidential prompt may be performed by, for example, Doc2Vec, the average of word embeddings, etc. For word embeddings, for example, Word2Vec, fastText, etc. are used.

[0145] The search unit 32 vectorizes the user setting information included in the search request in the same way as the vectorization of the heading information of the confidential prompt. Then, the search unit 32 acquires, as search results, the confidential prompts corresponding to the top m (m is 2 or more) pieces of heading information in descending order of the vector similarity with the user setting information being equal to or higher than a threshold value or having a high similarity from the confidential prompt table.

[0146] Note that the confidential prompt table may include information on each of a plurality of keywords included in the confidential prompt or its heading information instead of the vector information of the heading information of the confidential prompt. In this case, the search unit 32 acquires, as search results, the confidential prompts corresponding to the confidential prompts or heading information including one or more terms included in the user setting information included in the search request from the confidential prompt table.

[0147] Also, the confidential prompt table may include the vector information of the confidential prompt instead of the vector information of the heading information of the confidential prompt. In this case, the search unit 32 acquires, as search results, the top m (m is 2 or more) confidential prompts from the confidential prompt table in descending order of the vector similarity with the user setting information included in the search request being equal to or higher than a threshold value or having a high similarity.

[0148] [3.3.4. Selection unit 33] The selection unit 33 selects a confidential prompt corresponding to the request from the user U from among a plurality of confidential prompts that are a plurality of prompts with details hidden from the user U. For example, the selection unit 33 accepts the confidential prompt designated by the user U as the confidential prompt corresponding to the request from the user U.

[0149] For example, when the reception unit 30 receives a selection request including information for specifying one or more confidential prompts specified by the user U, the selection unit 33 selects one or more confidential prompts specified by the selection request from among a plurality of confidential prompts. The one or more confidential prompts specified by the selection request are the one or more confidential prompts specified by the user U. The selection unit 33 stores the identification information of the selected one or more confidential prompts in the storage unit 11.

[0150] In this case, the selection request includes information for specifying, for example, one or more confidential prompts specified by the user U from among a plurality of confidential prompts included in the search results displayed on the terminal device 2. Further, the selection request may include information for specifying one or more confidential prompts specified by the user U from among a plurality of confidential prompts included in the prompt list displayed on the terminal device 2 instead of the search results.

[0151] Also, when the reception unit 30 receives a selection request including information for specifying one or more selection prompt candidates specified by the user U, the selection unit 33 selects one or more selection prompt candidates specified by the selection request from among a plurality of selection prompt candidates.

[0152] The one or more selection prompt candidates specified by the selection request are the one or more selection prompt candidates specified by the user U. The selection unit 33 stores the identification information of the selected one or more selection prompt candidates in the storage unit 11. In this case, the selection request includes information for specifying, for example, one or more selection prompt candidates specified by the user U from among a plurality of selection prompt candidates displayed on the terminal device 2.

[0153] Also, when the reception unit 30 receives a provisional selection request, the selection unit 33 provisionally selects a plurality of confidential prompts specified by the provisional selection request from among a plurality of confidential prompts. The selection unit 33 stores the identification information of the provisionally selected plurality of confidential prompts in the storage unit 11.

[0154] Further, the selection unit 33 selects, as the privacy prompt corresponding to the user U's request, the privacy prompt specified by the user U from among the plurality of temporarily selected privacy prompts based on the selection information generated by the generation unit 34 based on the plurality of temporarily selected privacy prompts.

[0155] Also, when a usage request is received by the reception unit 30, the selection unit 33 selects one or more privacy prompts specified by the usage request from among the plurality of privacy prompts.

[0156] [3.3.5. Generation Unit 34] The generation unit 34 inputs information including the selection prompt, which is the privacy prompt selected by the selection unit 33, to the generation AI, and causes the generation AI to generate generation information, which is information corresponding to the selection prompt, which is the privacy prompt selected by the selection unit 33.

[0157] For example, assume that the selection prompt is a prompt that does not require additional information. In this case, the generation unit 34 inputs information including the selection prompt to the generation AI, and causes the generation AI to generate generation information, which is information corresponding to the selection prompt.

[0158] A selection prompt that does not require additional information is, for example, a prompt that collects information on a specific target whose state changes (e.g., a specific disaster, a specific person, a specific object) and instructs the generation of current information or summary information on the specific target.

[0159] In this case, the generation unit 34, for example, collects information on a specific target whose state changes (e.g., a specific disaster, a specific person, a specific object) according to the selection prompt, and inputs, as input information, information including the collected information and instruction information indicating an instruction to execute a specific process on the information, and causes the generation AI to generate generation information, which is information corresponding to the selection prompt, by the generation AI. The specific process is, for example, a process of summarizing the collected information in a specific format, a process of summarizing the collected information, etc., but is not limited to such examples.

[0160] A selection prompt that does not require additional information is a prompt that includes information of a character string that repeats a process of outputting a character string according to a predetermined condition, repeats a process of outputting an image according to a predetermined condition, or the like, or instructs a specific process.

[0161] Also, assume that the selection prompt is a prompt that requires additional information. In this case, the generation unit 34 inputs information including the information including the selection prompt and the additional information included in the selection request into the generation AI, and causes the generation AI to generate generation information that is information corresponding to the selection prompt and the additional information.

[0162] A selection prompt that requires additional information is, for example, a prompt that instructs execution of a specific task with respect to the additional information. The specific task is, for example, but not limited to, summarization of additional information, analysis of additional information (for example, sentiment analysis, etc.), translation of additional information, improvement or proofreading of additional information, generation of an article or story regarding additional information, and the like.

[0163] A selection prompt that requires additional information may be, for example, a prompt for executing a specific task in a chat format with the user U. In this case, the specific task is a task of performing Q&A (Question and Answer) regarding a specific category (for example, economy, education, etc.) in a chat format, and a task of inputting information indicating a question sentence to the user U as additional information and causing the generation AI to generate information indicating an answer sentence for the question sentence.

[0164] Also, the specific task may be a task of sequentially inputting information necessary for the user U as additional information for creation of information of a specific target (for example, various application documents, novels, cooking recipes, blogs, etc.), and causing the generation AI to generate information of the specific target based on these additional information.

[0165] The anonymized prompt selected as the selection prompt may be a prompt including intention definition information for extracting the intention type and intention content. The intention definition information includes, for example, instruction information including an instruction to extract the intention type and intention content from the user input information, and definition information including the intention type and information defining the intention type. When using the API provided by OpenAI, for example, the generation unit 34 can cause the generation AI to generate intention information by using the function of function calling.

[0166] The intention type is, for example, the type of intention of additional information, and the information indicating the intention type is, for example, information specifying a function or function corresponding to the intention type. Also, the intention content is the content of the intention of the additional information, and the information indicating the intention content is, for example, information indicating an argument of a function or a parameter of a function corresponding to the intention type.

[0167] In this case, intention information including information indicating the intention type and information indicating the intention content is generated by the generation AI. The generation unit 34 uses the intention information generated by the generation AI to obtain information from an external information processing device or an internal storage unit, etc., and inputs to the generation AI information including the obtained information and instruction information indicating an instruction to generate generated information using the information, and can cause the generation AI to generate generated information.

[0168] Also, when the selection prompt includes a message indicating a scenario, the generation unit 34 repeats the process of inputting information to the generation AI according to the scenario and obtaining an output until the end condition of the task is satisfied, and can generate generated information based on such a processing result. In this case, the generation unit 34 includes in the input information input to the generation AI an input / output history including information that was input to the generation AI in the past and information that was output by the generation AI in the past when executing the task.

[0169] For example, until the end condition of the task is satisfied, the generation unit 34 can cause the generation AI to generate generated information by inputting information including the input / output history to the generation AI as input information while changing the content to be generated by the generation AI according to the scenario.

[0170] The information input into the generative AI according to the scenario is information collected from an external information processing device or an internal storage unit, etc., or additional information input by the user U. For example, until the end condition of the task is satisfied, the generation unit 34 collects information from an external information processing device or an internal storage unit, etc. according to the scenario, and causes the generative AI to generate generated information based on the collected information.

[0171] Also, until the end condition of the task is satisfied, the generation unit 34 can also cause the user U to input additional information according to the scenario, and cause the generative AI to generate generated information based on this additional information.

[0172] When a plurality of anonymized prompts are selected by the selection unit 33, the generation unit 34 causes the generative AI to generate generated information for each anonymized prompt selected by the selection unit 33. Thereby, the generation unit 34 can obtain generated information for each selection prompt.

[0173] Also, the generation unit 34 inputs information including a selection prompt candidate, which is a prompt temporarily selected by the selection unit 33, into the generative AI, and causes the generative AI to generate candidate generated information, which is information corresponding to the selection prompt candidate, which is a prompt temporarily selected by the selection unit 33, for each selection prompt candidate. The method for generating the candidate generated information is the same as the method for generating the generated information described above.

[0174] 〔3.3.6. Providing Unit 35〕 The providing unit 35 provides various information to the user U. For example, the providing unit 35 provides various information to the user U by transmitting various information to the terminal device 2 via the communication unit 10 and the network N.

[0175] For example, when a prompt list request is received by the reception unit 30, the providing unit 35 provides the user U with prompt list information by transmitting to the terminal device 2 the prompt list information in which a plurality of anonymized prompts are listed so as to be individually selectable.

[0176] In addition, the providing unit 35 provides the user U with each of two or more anonymized prompts retrieved by the retrieval unit 32 in a selectable manner. For example, the providing unit 35 provides the user U with information on two or more anonymized prompts retrieved by the retrieval unit 32 by transmitting the information on two or more anonymized prompts retrieved by the retrieval unit 32 to the terminal device 2.

[0177] The terminal device 2 displays information on two or more anonymized prompts transmitted from the providing unit 35. The information on the anonymized prompts displayed on the terminal device 2 includes the title of the anonymized prompt and the heading information of the anonymized prompt, but does not include the content of the anonymized prompt itself. Therefore, the information processing apparatus 1 can provide the information on the anonymized prompt while keeping the detailed content of the anonymized prompt confidential. The user U confirms the titles and heading information of two or more anonymized prompts and selects one or more anonymized prompts from the two or more anonymized prompts displayed on the terminal device 2 by operating the terminal device 2.

[0178] In addition, the information on the anonymized prompts displayed on the terminal device 2 may include information indicating the charge amount of the anonymized prompt, information indicating the number of selections of the anonymized prompt, information indicating the evaluation of the anonymized prompt, and the like. In this case, the user U can confirm information indicating the charge amount of the anonymized prompt, information indicating the number of selections of the anonymized prompt, information indicating the evaluation of the anonymized prompt, and the like, and select one or more anonymized prompts from the two or more anonymized prompts displayed on the terminal device 2.

[0179] In addition, the providing unit 35 transmits the generated information generated by the generating unit 34 or the information based on the generated information to the terminal device 2 as providing information, thereby providing the user U with the generated information generated by the generating unit 34 or the information based on the generated information as providing information. Thereby, it is possible to provide the user U with the generated information generated using the anonymized prompt or the information based on the generated information as providing information while keeping the detailed content of the anonymized prompt confidential.

[0180] When the information generated by the generation unit 34 as generation information for the generation AI is the intended information, the providing unit 35 can acquire information from an external information processing device or the storage unit 11 or the like using such intended information, and generate providing information based on the acquired information.

[0181] In addition, when the generation information for each selection prompt is generated by the generation unit, the providing unit 35 provides the user U with the providing information for each selection prompt. In this case, the terminal device 2 displays the providing information for each selection prompt, and the user U can compare the providing information for each selection prompt.

[0182] The user U can check the providing information displayed on the terminal device 2, operate the terminal device 2, and input information indicating an evaluation of the providing information.

[0183] In addition, when the candidate generation information for each selection prompt candidate is generated by the process using the generation AI of the generation unit 34, the providing unit 35 provides the user U with selection information including the candidate providing information for each selection prompt candidate. For example, the providing unit 35 provides the selection information to the user U by transmitting the selection information to the terminal device 2.

[0184] The selection information includes, for each selection prompt candidate, candidate providing information that is candidate generation information or information based on the candidate generation information, associated with the information of the selection prompt candidate (for example, information indicating the title of the selection prompt candidate or heading information). The candidate providing information is the same type of information as the above-described providing information, but may be a part of the providing information.

[0185] In this case, the terminal device 2 displays the selection information including the candidate providing information for each selection prompt candidate, and the user U can compare the candidate providing information for each selection prompt candidate, and can specify and use a more appropriate confidentiality prompt.

[0186] The user U checks the candidate-providing information corresponding to each selection prompt candidate included in the selection information displayed on the terminal device 2, and designates one or more selection prompt candidates by operating the terminal device 2. In this case, the terminal device 2 transmits a selection request including information specifying one or more selection prompt candidates designated by the user U to the information processing device 1.

[0187] [3.3.7. Billing Unit 36] The billing unit 36 performs a billing process for the user U to whom the providing information has been provided by the providing unit 35. The billing process is, for example, a process of payment by the user U of the billing amount associated with the anonymized prompt used for generating the providing information provided to the user U in step S7.

[0188] The billing unit 36 can perform the billing process, for example, by transmitting billing information including information indicating the billing amount and information identifying the user U to the device of a financial institution, or by transmitting billing information including information indicating the billing amount and information identifying the credit card number of the user U to the device of a credit card company.

[0189] Also, instead of or in addition to a specific currency, the user U can pay the consideration for the anonymized prompt with points. For example, the billing unit 36 can also perform the billing process by transmitting billing information including information indicating the number of points corresponding to the billing amount and information identifying the user U to the device of a point management company.

[0190] Note that the points are, for example, rewards given to the user U by a point management company or the like when the user U shops with a credit card or cash, or when the user U subscribes to a specific service, but are not limited to such examples.

[0191] The charging unit 36 can perform a charging process for the user U to whom the provided information is provided by the providing unit 35 with different charging amounts depending on the use and purchase of the anonymization prompt. For example, in the case of purchasing an anonymization prompt, the charging process is performed at a lump-sum or monthly charging amount Ca for the provision of the provided information using the anonymization prompt. Also, in the case of using an anonymization prompt that the user U has not purchased, the charging process is performed at a per-use charging amount Cb (<Ca) for the provision of the provided information using the anonymization prompt.

[0192] [3.3.8. Decision unit 37] The decision unit 37 determines the charging amount for the use of the anonymization prompt. For example, the decision unit 37 determines the charging amount based on evaluation information, which is information indicating the evaluation of the provided information by each user U. The decision unit 37 determines a higher charging amount for an anonymization prompt with a higher average value of the evaluation of the provided information by each user U.

[0193] The decision unit 37 can determine the charging amount based on the evaluation of the provided information by each user U with reference to the reference amount set by the provider of the anonymization prompt. For example, the decision unit 37 determines a lower charging amount than the reference amount for an anonymization prompt with a lower average value of the evaluation of the provided information by each user U than the reference value. The reference value is 5 when the evaluation is a 10-point scale.

[0194] Also, the decision unit 37 can determine the charging amount based on the number of uses of the anonymization prompt by the user U. For example, the decision unit 37 determines a higher charging amount for an anonymization prompt with a larger number of uses of the anonymization prompt by the user U.

[0195] Also, the decision unit 37 can determine the charging amount based on the evaluation of the provided information by each user U and the number of uses of the anonymization prompt by the user U. For example, the decision unit 37 determines a higher charging amount for an anonymization prompt with a larger score obtained by weighted addition of the average value of the evaluation of the provided information by each user U and the number of uses of the anonymization prompt by the user U.

[0196] Further, the determination unit 37 can also determine the charging amount based on the number of purchases of the anonymized prompts. For example, the determination unit 37 determines a higher charging amount for anonymized prompts with a larger number of purchases.

[0197] Also, the determination unit 37 can determine the charging amount based on the evaluation of the provided information by each user U and the number of purchases of the anonymized prompts. For example, the determination unit 37 determines a higher charging amount for anonymized prompts with a larger score obtained by weighted addition of the average value of the evaluation of the provided information by each user U and the number of purchases of the anonymized prompts.

[0198] Also, the determination unit 37 can determine the charging amount based on the evaluation of the provided information by each user U, the number of uses of the anonymized prompts, and the number of purchases of the anonymized prompts. For example, the determination unit 37 determines a higher charging amount for anonymized prompts with a larger score obtained by weighted addition of the average value of the evaluation of the provided information by each user U, the number of uses of the anonymized prompts, and the number of purchases of the anonymized prompts.

[0199] [4. Processing Procedure] Next, the information processing procedure by the processing unit 12 of the information processing apparatus 1 according to the embodiment will be described. FIG. 6 is a flowchart showing an example of information processing by the processing unit 12 of the information processing apparatus 1 according to the embodiment.

[0200] As shown in FIG. 6, the processing unit 12 of the information processing apparatus 1 determines whether a search request has been received (step S20). When the processing unit 12 determines that a search request has been received (step S20: Yes), it searches for two or more anonymized prompts and provides the search results to the user U (step S21).

[0201] When the process in step S21 ends, or when it is determined that no search request has been received (step S20: No), the processing unit 12 determines whether a selection request has been received (step S22). When the processing unit 12 determines that a selection request has been received (step S22: Yes), it selects the confidential prompt specified in the selection request as the selection prompt, and generates generated information, which is information corresponding to the selection prompt, using the generation AI (step S23).

[0202] Then, the processing unit 12 provides the user U with the generated information or information based on the generated information as provided information (step S24). After that, the processing unit 12 performs a charging process for the user U to whom the provided information has been provided (step S25).

[0203] When the process in step S25 ends, or when it is determined that no selection request has been received (step S22: No), the processing unit 12 determines whether a provisional selection request has been received (step S26). When the processing unit 12 determines that a provisional selection request has been received (step S26: Yes), it generates candidate generated information using the generation AI (step S27). Then, the processing unit 12 provides the user U with the candidate generated information or information based on the candidate generated information as candidate provided information (step S28).

[0204] When the process in step S28 ends, or when it is determined that no provisional selection request has been received (step S26: No), the processing unit 12 determines whether any other request has been received (step S29). When the processing unit 12 determines that any other request has been received (step S29: Yes), it performs other processing (step S30).

[0205] When the process in step S30 ends, or when it is determined that no other request has been received (step S29: No), the processing unit 12 determines whether the operation end timing has been reached (step S31). The processing unit 12 determines that the operation end timing has been reached, for example, when the power of the information processing apparatus 1 is turned off.

[0206] When the processing unit 12 determines that the operation end timing has not been reached (step S31: No), the process proceeds to step S20. When it determines that the operation end timing has been reached (step S31: Yes), the process shown in FIG. 6 ends.

[0207] [[5. Modification Example]] In the above example, the user setting information included in the search request includes information indicating a string indicating operations, processes, etc. that the user U desires for the generated AI. Instead of or in addition to such information, information indicating the results that the user U desires for the generated AI may be included.

[0208] In this case, the confidential prompt table may further include, as information on the confidential prompt, generated information or provided information obtained using the generated AI from the confidential prompt and its vector information. In this case, the search unit 32 acquires, as search results, the confidential prompts corresponding to the top m (m is 2 or more) pieces of generated information or provided information in descending order of the vector similarity with the user setting information being equal to or higher than the threshold value or having a high similarity.

[0209] Further, the search unit 32 calculates a score obtained by weighted addition of the vector similarity between the user setting information and the heading information and the vector similarity between the user setting information and the generated information or provided information, and acquires, as search results, the confidential prompts corresponding to the combinations of the top m (m is 2 or more) pieces of heading information and generated information or generated information in descending order of such score being equal to or higher than the threshold value or having a high similarity from the confidential prompt table.

[0210] Further, the search unit 32 can also search for two or more confidential prompts from the confidential prompt table based on the attribute information, which is information indicating the attribute of the user U of the terminal device 2 that transmitted the search request, and the user setting information. The attribute of the user U is, for example, the demographic attribute or psychographic attribute of the user U.

[0211] For example, the search unit 32 can provide the user U with, as a search result, a confidential prompt having an attribute identical to or similar to the attribute of the user U of the terminal device 2 that sent the search request among a plurality of confidential prompts according to the user setting information, where the average value of evaluations by other users U having the same or similar attribute is equal to or higher than a threshold value or among the top m (m is 1 or more).

[0212] Further, the search unit 32 can provide the user U with, as a search result, a confidential prompt having an attribute identical to or similar to the attribute of the user U of the terminal device 2 that sent the search request among a plurality of confidential prompts according to the user setting information, where the number of uses by other users U having the same or similar attribute is equal to or higher than a threshold value or among the top m (m is 1 or more).

[0213] Also, when the confidential prompt is a prompt using a generation AI in a chat format, the determination unit 37 can determine the charging amount for the use of such a confidential prompt based on the input / output history performed in the chat format.

[0214] For example, the determination unit 37 can determine a higher charging amount for the use of such a confidential prompt as the number of input / outputs is larger or smaller, or can determine a higher charging amount for the use of such a confidential prompt as the number of tokens input as a character string to the generation AI is larger or smaller.

[0215] Also, the determination unit 37 can input, as input information to the generation AI, information including the input / output history and instruction information that is information of a character string indicating an instruction to determine the accuracy and value of the confidential prompt from such an input / output history, and cause the generation AI to output information indicating the accuracy and value of the confidential prompt. In this case, the determination unit 37 can determine a higher charging amount for the use of such a confidential prompt as the accuracy and value are higher.

[0216] Also, when an upper limit amount of the charging amount is set for each user U according to the request of the user U or the like, the determination unit 37 can also search for a confidential prompt with a charging amount that does not exceed the upper limit amount.

[0217] [6. Hardware Configuration] The information processing apparatus 1 according to the above-described embodiment is realized by a computer 80 having a configuration as shown in FIG. 7, for example. FIG. 7 is a hardware configuration diagram showing an example of a computer 80 that realizes the functions of the information processing apparatus 1 according to the embodiment. The computer 80 includes a CPU 81, a RAM 82, a ROM (Read Only Memory) 83, an HDD (Hard Disk Drive) 84, a communication interface (I / F) 85, an input / output interface (I / F) 86, and a media interface (I / F) 87.

[0218] The CPU 81 operates based on a program stored in the ROM 83 or the HDD 84 and controls each part. The ROM 83 stores a boot program executed by the CPU 81 when the computer 80 is started up, and programs dependent on the hardware of the computer 80.

[0219] The HDD 84 stores programs executed by the CPU 81 and data used by such programs. The communication interface 85 receives data from other devices via the network N (see FIG. 2) and sends it to the CPU 81, and sends data generated by the CPU 81 to other devices via the network N.

[0220] The CPU 81 controls output devices such as a display and a printer, and input devices such as a keyboard or a mouse via the input / output interface 86. The CPU 81 acquires data from the input device via the input / output interface 86. Also, the CPU 81 outputs data generated via the input / output interface 86 to the output device.

[0221] The media interface 87 reads a program or data stored in the recording medium 88 and provides it to the CPU 81 via the RAM 82. The CPU 81 loads such a program from the recording medium 88 onto the RAM 82 via the media interface 87 and executes the loaded program. The recording medium 88 is, for example, an optical recording medium such as a DVD (Digital Versatile Disc) or a PD (Phase change rewritable Disk), a magneto-optical recording medium such as an MO (Magneto-Optical disk), a tape medium, a magnetic recording medium, or a semiconductor memory.

[0222] For example, when the computer 80 functions as the information processing apparatus 1 according to the embodiment, the CPU 81 of the computer 80 realizes the functions of the processing unit 12 by executing the program loaded onto the RAM 82. Further, the data in the storage unit 11 is stored in the HDD 84. The CPU 81 of the computer 80 reads and executes these programs from the recording medium 88. As another example, these programs may be acquired from another device via the network N.

[0223] 〔7. Others〕 Also, among the respective processes described in the above embodiment, all or part of the processes described as being automatically performed can be manually performed, or all or part of the processes described as being manually performed can be automatically performed by a known method. In addition, the processing procedures, specific names, and information including various data and parameters shown in the above document and drawings can be arbitrarily changed unless otherwise specified. For example, the various information shown in each figure is not limited to the illustrated information.

[0224] Also, each component of each illustrated device is a functional concept, and does not necessarily have to be physically configured as illustrated. That is, the specific form of the distribution and integration of each device is not limited to that illustrated, and all or part of it can be functionally or physically distributed and integrated in any unit according to various loads and usage situations.

[0225] For example, the above-described information processing apparatus 1 may be implemented by a terminal device and a server computer, or may be implemented by a plurality of server computers. Also, depending on the function, the configuration can be flexibly changed, such as by calling an external platform or the like using an API or network computing.

[0226] In addition, the above-described embodiments and modifications can be appropriately combined as long as the processing contents do not conflict.

[0227] 〔8. Effect〕 As described above, the information processing apparatus 1 according to the embodiment includes a selection unit 33, a generation unit 34, and a provision unit 35. The selection unit 33 selects a prompt corresponding to a request from the user U from among a plurality of prompts whose details are concealed from the user U. The generation unit 34 inputs information including the prompt selected by the selection unit 33 to the generation AI, and causes the generation AI to generate information corresponding to the prompt selected by the selection unit 33. The provision unit 35 provides the user U with the information generated by the generation unit 34 or information based on the information as provision information. Thereby, the information processing apparatus 1 can perform information generation using a more appropriate prompt according to the request of the user U.

[0228] In addition, the information processing apparatus 1 includes a charging unit 36 that performs a charging process on the user U to whom the provision information is provided by the provision unit 35. Thereby, the information processing apparatus 1 can promote the provision of prompts according to the request of the user U, and can perform information generation using a more appropriate prompt according to the request of the user U.

[0229] In addition, the information processing apparatus 1 includes a determination unit 37 that determines a charging amount for the use of the prompt. Thereby, the information processing apparatus 1 can promote the provision of prompts according to the request of the user U.

[0230] Further, the determination unit 37 determines the charging amount based on the evaluation of the provided information by the user U. As a result, the information processing apparatus 1 can promote the provision of more appropriate prompts according to the requests of the user U.

[0231] Also, the generation unit 34 causes the generation AI to generate information corresponding to each of the plurality of prompts, and the provision unit 35 provides the user U with selection information including information corresponding to each of the plurality of prompts generated by the generation unit 34 or information based on each such information. The selection unit 33 selects, as a prompt requested by the user U, the prompt specified by the user U from among the plurality of prompts based on the selection information generated by the generation unit 34. As a result, the information processing apparatus 1 can more appropriately perform information generation using a more appropriate prompt according to the requests of the user U.

[0232] Also, the information processing apparatus 1 includes a reception unit 30 that receives a search request from the user U, and a search unit 32 that searches for two or more prompts corresponding to the search request received by the reception unit 30 as the plurality of prompts. The provision unit 35 provides each of the plurality of prompts searched by the search unit 32 to the user U in a selectable manner. As a result, the information processing apparatus 1 can more appropriately perform information generation using a more appropriate prompt according to the requests of the user U.

[0233] Also, the selection unit 33 selects, as a prompt requested by the user U, the prompt specified by the user U. As a result, the information processing apparatus 1 can more appropriately perform information generation using a more appropriate prompt according to the requests of the user U.

[0234] Also, the search unit 32 searches for the plurality of prompts using explanatory texts that explain the functions of the respective prompts. As a result, the information processing apparatus 1 can more appropriately perform information generation using a more appropriate prompt according to the requests of the user U.

[0235] As described above, the embodiments of the present application have been described in detail with reference to the drawings. However, this is merely an example, and the present invention can be implemented in other forms with various modifications and improvements based on the knowledge of those skilled in the art, including the aspects described in the column of the disclosure of the invention.

[0236] In addition, the "section (section, module, unit)" described above can be read as "means", "circuit", etc. For example, the acquisition unit can be read as an acquisition means or an acquisition circuit.

Description of Reference Numerals

[0237] 1 Information processing apparatus 2, 3 Terminal device 10 Communication unit 11 Storage unit 12 Processing unit 20 User information storage unit 21 Confidential prompt storage unit 30 Reception unit 31 Acquisition unit 32 Search unit 33 Selection unit 34 Generation unit 35 Provision unit 36 Billing unit 37 Decision unit 100 Information processing system N Network

Claims

1. A selection unit that selects a prompt according to the request from the user from among a plurality of prompts whose details are concealed from the user; A generation unit that inputs information including the prompt selected by the selection unit into a generation AI and causes the generation AI to generate information according to the prompt selected by the selection unit; A provision unit that provides the user with the information generated by the generation unit or information based on the information as provision information. An information processing apparatus characterized by the above.

2. A charging unit that performs a charging process on the user to whom the provision information is provided by the provision unit. The information processing apparatus according to claim 1, characterized by the above.

3. A determination unit that determines the amount to be charged for the use of the prompt. The information processing apparatus according to claim 2, characterized by the above.

4. The determination unit: Determines the amount to be charged based on the evaluation by the user of the provision information. The information processing apparatus according to claim 3, characterized by the above.

5. The generation unit: Causes the generation AI to generate information according to each of the plurality of prompts; The provision unit: Provides the user with selection information including the information according to each of the plurality of prompts generated by the generation unit or information based on each of the pieces of information; The selection unit: Selects, from among the plurality of prompts, the prompt designated by the user as the prompt according to the user's request based on the selection information generated by the generation unit. The information processing apparatus according to any one of claims 1 to 4, characterized by the above.

6. A reception unit that receives a search request from the user; A search unit that searches for two or more prompts according to the search request received by the reception unit as the plurality of prompts; The provision unit: Provides the user with each of the plurality of prompts searched by the search unit in a selectable manner. The information processing apparatus according to claim 5, characterized by the above.

7. The selection unit: Selects the prompt designated by the user as the prompt according to the user's request. The information processing apparatus according to any one of claims 1 to 4, characterized by the above.

8. The search unit: Searches for the plurality of prompts using a description text that explains the function of each prompt. The information processing apparatus according to claim 6, characterized by the above.

9. An information processing method executed by a computer, comprising: a selection step of selecting a prompt corresponding to a request from the user from among a plurality of prompts whose details are concealed from the user; a generation step of inputting information including the prompt selected in the selection step into a generation AI and causing the generation AI to generate information corresponding to the prompt selected in the selection step; a provision step of providing the user with the information generated in the generation step or information based on the information as provision information. An information processing method characterized by the above.

10. a selection procedure of selecting a prompt corresponding to a request from the user from among a plurality of prompts whose details are concealed from the user; a generation procedure of inputting information including the prompt selected in the selection procedure into a generation AI and causing the generation AI to generate information corresponding to the prompt selected in the selection procedure; a provision procedure of causing a computer to execute a provision procedure of providing the user with the information generated in the generation procedure or information based on the information as provision information. An information processing program characterized by the above.

Images