Information processing systems, information processing methods, and computer programs

Abstract

This technology provides a way to improve the convenience of user interaction with multiple artificial intelligences. [Solution] Generated sentences are obtained from each of multiple artificial intelligences. The generated sentences obtained from each of the multiple artificial intelligences are displayed on a display device. User input sentences are obtained through an input device and sent to at least one of the multiple artificial intelligences. The generated sentences include a response sentence to the input sentence and are generated by the corresponding artificial intelligence. When a first instruction is entered from the user through the input device, the input sentence is sent to one of the multiple artificial intelligences specified by the user. When a second instruction is entered from the user through the input device, the input sentence is sent to multiple artificial intelligences.

Description

Technical Field

[0006] ,

[0001] This disclosure relates to an information processing system and an information processing method.

Background Art

[0002] Conventionally, a dialogue system that realizes dialogue with a user has been known (see, for example, Patent Document 1). The dialogue is performed, for example, through text or voice. As a conventional system, a system using artificial intelligence is also known. Artificial intelligence is based on, for example, a large language model. As artificial intelligence, ChatGPT (registered trademark) is widely known.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] In the conventional technology for realizing dialogue using artificial intelligence, an information processing system that mediates between the user and the artificial intelligence transmits an input sentence by the user to the artificial intelligence. The artificial intelligence generates a response sentence based on the input sentence. The information processing system acquires the response sentence generated by the artificial intelligence and displays the response sentence to the user through a user interface.

[0005] The conventional system is a simple system that realizes one-to-one dialogue between the user and the artificial intelligence. Therefore, according to one aspect of the present disclosure, it is desirable to provide a highly convenient technology related to dialogue between a user and a plurality of artificial intelligences.

Means for Solving the Problems

[0006] According to one aspect of this disclosure, an information processing system is provided. The information processing system comprises a generated text acquisition unit, a display control unit, and a transmission unit. The generated text acquisition unit is configured to acquire generated texts from each of a plurality of artificial intelligences.

[0007] The display control unit is configured to display the generated sentences obtained from each of the multiple artificial intelligences on a display device. The transmission unit is configured to obtain input sentences from the user through an input device and transmit the input sentences to at least one of the multiple artificial intelligences.

[0008] The generated sentence includes a response sentence to the input sentence and is generated by the corresponding artificial intelligence. When the first instruction is input from the user via the input device, the transmission unit sends the input sentence to one of the multiple artificial intelligences specified by the user. When the second instruction is input from the user via the input device, the transmission unit sends the input sentence to multiple artificial intelligences.

[0009] With the information processing system configured in this way, a user can send a common input message to multiple artificial intelligences. According to one aspect of this disclosure, it is possible to provide an information processing system that is highly convenient for users to use multiple artificial intelligences.

[0010] According to one aspect of this disclosure, when a third instruction is input from the user through an input device, the transmission unit transmits the content of the dialogue between the user and the first artificial intelligence among the multiple artificial intelligences, through the input sentence and generated sentence, to a second artificial intelligence that is different from the first artificial intelligence among the multiple artificial intelligences. It may be configured to transmit to the artificial intelligence. The generated sentence acquisition unit may be configured to acquire a response sentence to the dialogue content as a generated sentence from the second artificial intelligence.

[0011] With an information processing system configured in this way, a user can, for example, obtain a response from another artificial intelligence to the content of their conversation with an artificial intelligence. For example, a user can obtain a different idea from a second artificial intelligence that incorporates the ideas of the first artificial intelligence, making it possible to make meaningful use of multiple artificial intelligences.

[0012] According to one aspect of this disclosure, the first artificial intelligence and the second artificial intelligence may each be an artificial intelligence designated by the user from among several artificial intelligences. That is, the first artificial intelligence and the second artificial intelligence may each be designated by the user. In this case, the user can meaningfully utilize the functions realized by the third instruction described above by designating the artificial intelligence.

[0013] According to one aspect of this disclosure, the transmitting unit may be configured to selectively transmit a portion of the dialogue between the user and the first artificial intelligence to the second artificial intelligence when a third instruction is input. By providing the second artificial intelligence with only a portion of the dialogue rather than the entirety, it is possible to obtain a response from the second artificial intelligence that focuses on that portion of the dialogue.

[0014] According to one aspect of this disclosure, when a fourth instruction is input from a user through an input device, the transmission unit may be configured to cause the first artificial intelligence and the second artificial intelligence to interact by transmitting a generated sentence obtained from the first artificial intelligence among the multiple artificial intelligences to a second artificial intelligence, which is different from the first artificial intelligence among the multiple artificial intelligences, and transmitting a generated sentence obtained from the second artificial intelligence to the first artificial intelligence.

[0015] With this type of information processing system, users can view the results of discussions between artificial intelligences regarding a theme or topic of interest. These discussion results can, for example, aid in the user's deeper thinking about the theme or topic of interest.

[0016] According to one aspect of this disclosure, the first artificial intelligence and the second artificial intelligence may each be an artificial intelligence designated by the user from among several artificial intelligences. That is, the first artificial intelligence and the second artificial intelligence may each be designated by the user. In this case, the user can meaningfully utilize the functions realized by the fourth instruction above by designating the artificial intelligence.

[0017] According to one aspect of this disclosure, the transmission unit may be configured to transmit, when a fifth instruction is input from a user through an input device, a generated sentence obtained from a first artificial intelligence among a plurality of artificial intelligences to a second artificial intelligence among the plurality of artificial intelligences that is different from the first artificial intelligence, and a generated sentence obtained from the second artificial intelligence to a third artificial intelligence among the plurality of artificial intelligences that is different from the first and second artificial intelligences.

[0018] With this type of information processing system, users can obtain the results of the combined thoughts of multiple artificial intelligences from a third artificial intelligence. According to one aspect of this disclosure, the transmitting unit may transmit the content of the conversation between the user and each of the multiple artificial intelligences to a specific artificial intelligence that is different from the multiple artificial intelligences. The transmitting unit may transmit the input text to a specific artificial intelligence when a sixth instruction is input from the user through an input device.

[0019] The generated text acquisition unit may acquire a generated text from a specific artificial intelligence based on an input text sent to that specific artificial intelligence. The display control unit displays the generated text acquired from the specific artificial intelligence on a display device. It may be configured to display on the screen.

[0020] With an information processing system configured in this way, a user can obtain a response from a specific artificial intelligence that comprehensively considers the user's interactions with multiple artificial intelligences.

[0021] According to one aspect of the present disclosure, the plurality of artificial intelligences may include two or more artificial intelligences that play different roles. According to the information processing system configured in this way, a user can receive the provision of meaningful information from the plurality of artificial intelligences.

[0022] According to one aspect of the present disclosure, the two or more artificial intelligences may include an artificial intelligence that plays the role of a devil's advocate. According to the information processing system configured in this way, a user can realize deep thinking about various themes through dialogue with the devil's advocate.

[0023] According to one aspect of the present disclosure, the display control unit may control the display device to display a display screen including a plurality of partitioned areas on the display device, and cause the display device to display a generated sentence generated by a corresponding one of the plurality of artificial intelligences in each of the plurality of areas on the display screen. According to the information processing system configured in this way, the content of the dialogue between the user and the plurality of artificial intelligences can be displayed on the display screen in a form that is easily understandable by the user.

[0024] According to one aspect of the present disclosure, the display control unit may be configured to receive a user operation on a plurality of areas through an input device and change the order of the plurality of areas on the display screen according to the user operation. According to the information processing system configured in this way, a display screen that is easier for the user to understand regarding the content of the dialogue between the user and the plurality of artificial intelligences can be displayed for the user.

[0025] According to one aspect of the present disclosure, an input field for writing an input sentence for a corresponding one of the plurality of artificial intelligences may be displayed in each of the plurality of areas.

[0026] According to one aspect of the present disclosure, the transmission unit may be configured to transmit the dragged generated sentence to the artificial intelligence corresponding to the input field where the generated sentence is dropped on the condition that the generated sentence displayed in the first area among the plurality of areas is dragged and dropped into the input field of the second area among the plurality of areas.

[0027] According to one aspect of this disclosure, the generated sentence acquisition unit may be configured to acquire a response sentence to a dragged generated sentence from artificial intelligence corresponding to the input field where the generated sentence was dropped.

[0028] With an information processing system configured in this way, users can obtain responses from other artificial intelligences to the output of one artificial intelligence, enabling them to make meaningful use of multiple artificial intelligences.

[0029] According to one aspect of this disclosure, each of the multiple areas may display the content of a conversation between a user and a corresponding artificial intelligence by arranging the user's input sentence and the corresponding artificial intelligence's generated sentence.

[0030] According to one aspect of this disclosure, the transmission unit may be configured to transmit the dragged generated sentence, along with a portion of the dialogue that describes the context in which the dragged generated sentence was generated, to the artificial intelligence corresponding to the input field where the generated sentence was dropped.

[0031] According to this information processing system, the user can, with simple operations, interact with a second person. It is possible to obtain appropriate responses from artificial intelligence.

[0032] According to one aspect of this disclosure, the display control unit may be configured to control the display device so as to display a plurality of display windows on the display device, and to display a generated sentence generated by a corresponding artificial intelligence among a plurality of artificial intelligences in each of the plurality of display windows.

[0033] The transmitting unit may be configured to determine that a fourth instruction has been input, provided that an operation is performed via an input device to overlay the display window associated with the first artificial intelligence onto the display window associated with the second artificial intelligence, and to cause the first and second artificial intelligences to interact.

[0034] With an information processing system configured in this way, it is possible to improve the user's operability regarding the input of the fourth instruction.

[0035] According to one aspect of this disclosure, for processing based on the fifth instruction, the transmission unit may be configured as follows: The transmission unit may be configured to receive additional instructions from a user via an input device for at least one of the second artificial intelligence and the third artificial intelligence. In this case, the transmission unit may be configured to input the additional instructions received for the second artificial intelligence to the second artificial intelligence when transmitting the generated text obtained from the first artificial intelligence to the second artificial intelligence. The transmission unit may be configured to input the additional instructions received for the third artificial intelligence to the third artificial intelligence when transmitting the generated text obtained from the second artificial intelligence to the third artificial intelligence.

[0036] With this type of information processing system, users can control the direction of thought by inputting additional instructions, and obtain the result of the combined thoughts of multiple artificial intelligences from a third artificial intelligence.

[0037] According to one aspect of this disclosure, for processing based on the fourth instruction, the transmission unit may be configured as follows: When the fourth instruction is input, the transmission unit may cause the first artificial intelligence and the second artificial intelligence to converse until at least one of the generated sentences obtained from the first artificial intelligence and the generated sentences obtained from the second artificial intelligence among the plurality of artificial intelligences satisfies a predetermined condition, and then terminate the conversation between the first artificial intelligence and the second artificial intelligence once the predetermined condition is satisfied.

[0038] With this configured information processing system, the dialogue between artificial intelligences can be terminated when predetermined conditions are met, without requiring user input for termination. Therefore, this information processing system improves user convenience.

[0039] According to one aspect of this disclosure, one of the first artificial intelligence and the second artificial intelligence may be configured to score the generated sentence input from the other of the first and second artificial intelligences. The transmission unit may be configured to determine that a predetermined condition has been met when the score assigned to the generated sentence by scoring is equal to or greater than a threshold, and to terminate the dialogue between the first artificial intelligence and the second artificial intelligence.

[0040] According to one aspect of this disclosure, a computer program for causing a computer to function may be provided as at least a part of the generated text acquisition unit, display control unit, and transmission unit of the information processing system described above.

[0041] According to one aspect of this disclosure, an information processing method corresponding to the information processing system described above is provided. The information processing method may be performed by a computer.

[0042] According to one aspect of this disclosure, the information processing method may include obtaining generated text from each of several artificial intelligences. The information processing method may also include displaying the generated text obtained from each of the several artificial intelligences on a display device.

[0043] The information processing method may include obtaining an input sentence from a user through an input device and transmitting the input sentence to at least one of several artificial intelligences. The generated sentence includes a response sentence to the input sentence and may be generated by the corresponding artificial intelligence.

[0044] Sending may include, when a first instruction is input from the user through an input device, sending the input sentence to one of several artificial intelligences designated by the user, and when a second instruction is input from the user through an input device, sending the input sentence to multiple artificial intelligences.

[0045] This information processing method, configured in this way, produces the same effects as the information processing system described above. [Brief explanation of the drawing]

[0046] [Figure 1] This is a block diagram showing the configuration of the information processing system of the first embodiment. [Figure 2] This is a diagram (part 1) illustrating the configuration of the service provision screen. [Figure 3] This is a diagram (part 2) illustrating the configuration of the service provision screen. [Figure 4] Figure 4A is an explanatory diagram regarding the transmission of AI input text when the send button is pressed, and Figure 4B is an explanatory diagram regarding the transmission of AI input text when the mass send button is pressed. [Figure 5] This is an explanatory diagram regarding rearranging the chat screen. [Figure 6] This is an explanatory diagram regarding the drag-and-drop operation of spoken text. [Figure 7] This diagram illustrates the transmission of a spoken message to another artificial intelligence. [Figure 8] This is an explanatory diagram regarding the superposition operation. [Figure 9] This diagram illustrates the sending and receiving of AI input and output sentences in AI-to-AI dialogue. [Figure 10] This is an explanatory diagram regarding the cascade function. [Figure 11] This is a flowchart (part 1) showing the operation acceptance process executed by the processor. [Figure 12] This is a flowchart (part 2) illustrating the operation acceptance process executed by the processor. [Figure 13] This is a flowchart representing the AI-to-AI dialogue process executed by the processor. [Figure 14] This is a flowchart illustrating the cascading-related processes performed by the processor. [Figure 15] This is a flowchart illustrating the master-related processes performed by the processor. [Figure 16] This is an explanatory diagram regarding the generation of artificial intelligence with a scoring function. [Figure 17] This diagram illustrates an example of AI output text with scoring capabilities. [Figure 18] This is a flowchart representing the termination control process executed by the processor. [Figure 19] This is an explanatory diagram regarding additional instructions. [Figure 20] This diagram illustrates the user interface for additional instructions. [Figure 21] This is a flowchart illustrating the relay-related processing performed by the processor. [Modes for carrying out the invention]

[0047] Exemplary embodiments of the present disclosure are described below with reference to the drawings. [First Embodiment] The information processing system 10 of this embodiment, shown in Figure 1, communicates with an information terminal 60 and an AI server 70 to provide a user operating the information terminal 60 with a dialogue service with multiple artificial intelligences 71. It is configured to provide multiple artificial intelligences 71. The AI ​​server 70 functions as multiple artificial intelligences 71.

[0048] Multiple artificial intelligences 71 include, for example, two or more artificial intelligences 71 each with different virtual personalities. In other words, multiple artificial intelligences 71 include two or more artificial intelligences 71 each playing different roles.

[0049] Multiple artificial intelligences 71 may include two or more artificial intelligences 71 with the same configuration. Multiple artificial intelligences 71 may include a mixture of two or more artificial intelligences 71 with different configurations and two or more artificial intelligences 71 with the same configuration.

[0050] For example, multiple artificial intelligences 71 may include two or more artificial intelligences 71 that portray people of different age groups and genders. For example, multiple artificial intelligences 71 may include two or more artificial intelligences 71 that portray people of different occupations, qualifications, and / or expertise. For example, multiple artificial intelligences 71 may include two or more artificial intelligences that function as writers.

[0051] The information processing system 10 can prepare multiple artificial intelligences 71 on the AI ​​server 70 by inputting instructions for creating artificial intelligences 71, which include information specifying the configuration of the artificial intelligences 71, into the AI ​​server 70.

[0052] The information specifying the configuration of artificial intelligence 71 includes instructions for artificial intelligence 71 and the knowledge to be given to artificial intelligence 71. These instructions specify what behavior artificial intelligence 71 should exhibit, in other words, what role it should play. By providing knowledge to artificial intelligence 71, it is possible to make artificial intelligence 71 perform behaviors based on that specific knowledge.

[0053] As shown in Figure 1, the information processing system 10 includes a processor 11, memory 13, storage 15, and a communication interface 19. The processor 11 is configured to execute processing according to a computer program. The memory 13 functions as working memory when the processor 11 is executing processing. The memory 13 is, for example, RAM.

[0054] The storage 15 stores computer programs as well as various data used for processing performed by the processor 11. An example of the storage 15 is a solid-state drive (SSD).

[0055] The communication interface 19 is configured to communicate with the information terminal 60 and the AI ​​server 70 via a wide-area network (e.g., the Internet). In Figure 1, only one information terminal 60 is shown as a simplified representation. However, the information processing system 10 can be configured to communicate with multiple information terminals 60 corresponding to multiple users.

[0056] The information terminal 60 is, for example, an information terminal owned by the user. Examples of information terminals 60 include personal computers, smartphones, and tablet terminals. The information terminal 60 includes a user interface 61. The user interface 61 includes a display device 61A and an input device 61B.

[0057] The display device 61A is configured to display information for the user. An example of the display device 61A is a liquid crystal display. The input device 61B is configured to accept user input. An example of the input device 61B is a keyboard, a pointing device, and a touch panel.

[0058] The information terminal 60 accesses the information processing system 10 in accordance with user operations via the input device 61B and communicates with the information processing system 10 regarding the interactive service. At this time, the information processing system 10 displays the service provision screen G0 on the display device 61A of the information terminal 60.

[0059] The information processing system 10 enables interaction between the user and multiple artificial intelligences 71 by receiving user operations on the service provision screen G0 through the input device 61B.

[0060] The information processing system 10 receives input text for the artificial intelligence 71 through the input device 61B and the service provision screen G0. Hereinafter, the input text for the artificial intelligence 71 will be referred to as AI input text. The information terminal 60 transmits the AI ​​input text entered to the service provision screen G0 via the input device 61B to the information processing system 10.

[0061] The information processing system 10 transmits the AI ​​input sentence received from the information terminal 60 to at least one of the multiple artificial intelligences 71. Of the multiple artificial intelligences 71, the one that receives the AI ​​input sentence generates a response sentence to the AI ​​input sentence and sends it back to the information processing system 10. Hereinafter, the generated sentence produced by the artificial intelligence 71, including the response sentence, will be referred to as the AI ​​output sentence.

[0062] The information processing system 10 displays the received AI output text on the display device 61A of the information terminal 60, specifically on the service provision screen G0. This enables interaction between the user and multiple artificial intelligences 71 through the service provision screen G0.

[0063] As shown in Figure 2, the service provision screen G0 is configured to display AI output sentences from multiple artificial intelligences 71, and is also configured to accept input operations for AI input sentences to multiple artificial intelligences 71.

[0064] The service provision screen G0 includes multiple chat screens G1 and G3 corresponding to multiple artificial intelligences 71. According to this embodiment, multiple artificial intelligences 71 are designated by the user as conversation partners via the input device 61B. The service provision screen G0 displays the chat screens G1 and G3 related to the multiple artificial intelligences 71 designated as conversation partners.

[0065] In this embodiment, the multiple artificial intelligences 71 that the user can designate as conversation partners include a specific artificial intelligence 71M (hereinafter referred to as Master 71M) that has been assigned a specific role.

[0066] Master 71M is an artificial intelligence (AI) assigned the role of overseeing the other AIs 71 among the multiple AIs 71. Hereafter, each of the other AIs 71 will be referred to as AI 71A, using the code 71A. As mentioned above, the multiple AIs 71A include two or more AIs 71A playing different roles.

[0067] Figure 2 shows the configuration of the service provision screen G0 when the user designates three artificial intelligences 71A and one master 71M as conversation partners. The chat screen G3 shown in Figure 2 is the chat screen associated with master 71M. Each of the multiple chat screens G1 is a chat screen associated with one of the multiple artificial intelligences 71A other than master 71M.

[0068] The chat screen G1 comprises a profile image G11, a chat display screen G13, an input field G15, and a send button G18. The profile image G11 is associated with This is an area on the chat screen G1 that displays an anthropomorphic image of artificial intelligence 71 and the name assigned to artificial intelligence 71A. For example, if artificial intelligence 71A is playing the role of a woman in her 20s, the profile image G11 will display an avatar image of a woman in her 20s and the name of artificial intelligence 71A that allows the user to identify that she is a woman in her 20s.

[0069] The chat display screen G13 is a vertically scrolling screen. The chat display screen G13 displays the content of the conversation between the user and the artificial intelligence 71A, arranged chronologically from top to bottom using speech bubbles. The input field G15 is configured to accept text input from the user via the input device 61B.

[0070] The send button G18 is an operation object that accepts instructions to send text entered in the input field G15 of the same chat screen G1 to the artificial intelligence 71A. When the send button G18 is pressed, the text written in the input field G15 of the same chat screen G1 is sent to the corresponding artificial intelligence 71A as the AI ​​input message.

[0071] The send button G18 temporarily changes to the mass send button G19 when a specific operation is performed via the input device 61B. Figure 3 shows the configuration of the service provision screen G0 equipped with the mass send button G19. For example, the send button G18 changes to the mass send button G19 only while a specific key on the input device 61B (e.g., the shift key on a keyboard) is pressed.

[0072] The mass send button G19 is an operation object that receives a mass send instruction, which instructs the system to send the text entered in the input field G15 of the same chat screen G1 as an AI input message to all of the multiple artificial intelligences 71A whose chat screen G1 is displayed on the service provision screen G0.

[0073] In other words, pressing the send button G18 sends the AI ​​input text to a single corresponding artificial intelligence 71A, as shown in Figure 4A, whereas pressing the simultaneous send button G19 sends the AI ​​input text to multiple artificial intelligences 71A simultaneously, as shown in Figure 4B. In this respect, the send button G18 and the simultaneous send button G19 are different.

[0074] Next, the configuration of the chat screen G3, which is associated with the master 71M, will be explained. As shown in Figures 2 and 3, the chat screen G3 comprises a profile image G31, a chat display screen G33, an input field G35, and a group of operation buttons G37.

[0075] The profile image G31 is an area on the chat screen G3 that displays an anthropomorphic image of the associated master 71M, along with the name assigned to master 71M. The chat display screen G33 is a vertically scrolling screen.

[0076] The chat display screen G33 shows the conversations between multiple artificial intelligences 71, including Master 71M, and the user, arranged chronologically from top to bottom using speech bubbles. The chat display screen G33 also shows the conversations between multiple artificial intelligences 71A and the user, along with information about the speaker.

[0077] The input field G35 is configured to accept text input from the user via the input device 61B. The operation button group G37 includes a send button G371, a call button G373, a notepad button G375, and an output button G379.

[0078] The send button G371 is an operation object that accepts the instruction to send the text entered in input field G35 to master 71M. The send button G371 is pressed. The text entered in input field G35 is then sent as AI input to master 71M, which is associated with chat screen G3.

[0079] The call button G373 is a button used by the user to specify the person they wish to talk to. When the call button G373 is pressed, a selection screen for artificial intelligences 71 (not shown) is displayed. The user can, via the input device 61B, specify one or more artificial intelligences 71A and one master 71M from among the multiple artificial intelligences 71 displayed on the selection screen as the person they wish to talk to. When a person is specified via the call button G373, the chat screen G1 for each of the specified artificial intelligences 71A is displayed on the service provision screen G0.

[0080] The chat screen G3 is displayed on the service provision screen G0 even before a master 71M is designated as the conversation partner. For example, a standard master 71M is prepared in advance as the conversation partner without any user selection.

[0081] The notepad button G375 is used when the user takes notes via the input device 61B. When the notepad button G375 is pressed, an electronic notepad is displayed on the display device 61A. The information written in the notepad is recorded in the information processing system 10 and is not transmitted to the artificial intelligence 71.

[0082] The output button G379 is an operation object that receives instructions to output the content of the conversation between the user and the artificial intelligence 71. For example, when the output button G379 is pressed, the information processing system 10 is configured to output the content of the conversation between the user and multiple artificial intelligences 71 as text. The information processing system 10 may also be configured to output the content of the conversation between the user and multiple artificial intelligences 71 as a summary of meeting minutes as text. The meeting minutes can be generated by the master 71M.

[0083] As shown in Figure 5, each of the chat screens G1 and G3 within the service provision screen G0 functions as an individual display window. The service provision screen G0 is configured so that the left-right arrangement of each chat screen G1 and G3 (display window) can be changed by the user through the input device 61B to move the chat screens G1 and G3 (display windows). The movement of the chat screens G1 and G3 (display windows) can be achieved, for example, by dragging.

[0084] Furthermore, the information processing system 10 is configured to transmit, with respect to multiple artificial intelligences 71A, a statement selected by the user from the content of the conversation between the first artificial intelligence and the user to a second artificial intelligence, which is separate from the first artificial intelligence. The service provision screen G0 is equipped with a graphical user interface for this purpose. Here, the first and second artificial intelligences are arbitrarily selected by the user from among multiple artificial intelligences 71A designated as conversation partners.

[0085] As shown in Figure 6, the AI ​​input text and AI output text displayed in the form of speech bubbles on the chat screen G1 in the service provision screen G0 are each independently configured as drag-and-drop operable objects OJ1. Hereafter, the term "speech text" will be used to refer to both AI input text and AI output text. For example, the AI ​​input text is the user's speech text, and the AI ​​output text is the speech text of the artificial intelligence 71. However, as will be described later, the AI ​​input text may also be the speech text of the artificial intelligence 71.

[0086] On the service provision screen G0, the user drags a speech bubble (operation object OJ1) containing the message they want to send to the second artificial intelligence from the chat display screen G13 of the chat screen G1 corresponding to the first artificial intelligence, and inputs it into the chat screen G1 corresponding to the second artificial intelligence. You can perform a drag-and-drop operation by dropping items into column G15.

[0087] The information processing system 10 is configured to transmit a speech bubble specified by the user (hereinafter referred to as the "transferred text") to the second artificial intelligence through this drag-and-drop operation (hereinafter referred to as the "transfer operation").

[0088] This transfer operation copies the transferred message in a modified form into input field G15 of the chat screen G1, which corresponds to the second artificial intelligence. The modified transferred message has the speaker's information corresponding to the transferred message appended to the end of the original message.

[0089] When the forwarded message is an AI output message from the first artificial intelligence, the identification information of the first artificial intelligence is written at the end of the processed forwarded message as information about the speaker. When the name of the speaker (for example, the first artificial intelligence) is "〇〇", the special string "[〇〇's opinion]" is written at the end of the processed forwarded message.

[0090] When the processed transfer text is entered in input field G15 and the send button G18 is pressed, the processed transfer text is sent to the second artificial intelligence. As an alternative, the information processing system 10 may be configured to send the processed transfer text to the second artificial intelligence as soon as the transfer text is dropped into input field G15, without waiting for the send button G18 to be pressed.

[0091] More specifically, as shown in Figure 7, the information processing system 10 generates an AI input sentence (hereinafter referred to as "dialogue input sentence") D3 for the second artificial intelligence by adding a series of utterances D2 that lead to the processed transfer sentence D1 to the processed transfer sentence D1, and transmits this dialogue input sentence D3 to the second artificial intelligence.

[0092] Figure 7 illustrates how the information processing system 10 generates a dialogue input sentence D3 by attaching a series of utterances D2 to the processed transfer sentence D1, and transmits this to the second artificial intelligence. As can be seen from Figure 7, in the dialogue input sentence D3, each of the utterances D21 included in the series of utterances D2 is accompanied by speaker information. The speaker information is information that allows the artificial intelligence 71 to identify the speaker corresponding to the utterance D21.

[0093] The series of statements D2 is sent to the second artificial intelligence to convey to the second artificial intelligence the context (i.e., the background) in which the statements corresponding to the forwarded statements occurred. The series of statements D2 corresponds to a part of the dialogue between the user and the first artificial intelligence. In other words, the series of statements D2 is a partial extraction of the dialogue content related to the forwarded statements from the dialogue between the user and the first artificial intelligence. The series of statements D2 is composed of a predetermined number of statements D21 included in the past continuous dialogue between the user and the first artificial intelligence leading up to the forwarded statements, arranged in chronological order.

[0094] A distinctive feature of this embodiment is that the dialogue input statement D3 also stores a series of utterances between the user and the second artificial intelligence as related statement D4. Each of the utterances D41 included in related statement D4 is accompanied by speaker information. Related statement D4 is constructed by arranging a series of utterances in chronological order, with the most recent utterance in the dialogue between the user and the second artificial intelligence as the endpoint. Specifically, it is constructed by arranging a predetermined number of past utterances D41, including the most recent utterance, that occurred in the dialogue between the user and the second artificial intelligence, in chronological order.

[0095] When such dialogue input sentence D3 is input to the second artificial intelligence, a meaningful response to the processed transfer sentence D1 from the second artificial intelligence is expected, taking into account the content of the dialogue between the first artificial intelligence and the user, and the content of the dialogue between the second artificial intelligence and the user.

[0096] According to this embodiment, each artificial intelligence 71A contains the special string "[Opinion of XX]". When this special string is included in the AI ​​input text, the AI ​​is instructed to provide a response to the utterance (transferred text) associated with this special string, taking into account the context leading up to the utterance. Such instructions enable a meaningful response from the second artificial intelligence to the processed transfer text D1.

[0097] In addition, the information processing system 10 is configured to enable dialogue between artificial intelligences 71A (hereinafter referred to as "AI-to-AI dialogue"). This AI-to-AI dialogue is available when two or more chat screens G1 of artificial intelligences 71A are displayed on the service provision screen G0.

[0098] In this state, the user can select a first artificial intelligence and a second artificial intelligence from among two or more artificial intelligences 71A to be the target of inter-AI dialogue, and perform an operation (hereinafter referred to as "overlay operation") to move the chat screen G1 of the first artificial intelligence by dragging it and overlaying it onto the chat screen G1 of the second artificial intelligence, as shown in Figure 8.

[0099] When the chat screen G1 of the first artificial intelligence and the chat screen G2 of the second artificial intelligence are superimposed by this superposition operation, the information processing system 10 determines that an instruction has been given to start an AI-to-AI dialogue between the first artificial intelligence and the second artificial intelligence.

[0100] Subsequently, as shown in Figure 9, the information processing system 10 sends the AI ​​output sentence of the first artificial intelligence to the second artificial intelligence as an AI input sentence, and sends the AI ​​output sentence from the second artificial intelligence as a response sentence to this AI input sentence to the first artificial intelligence as an AI input sentence. This process is repeated until an instruction to terminate the AI-to-AI dialogue is received.

[0101] This AI-to-AI dialogue begins after a superposition operation, when user input text is sent as an AI input sentence to either the first or second artificial intelligence, and the dialogue starts with an AI output sentence produced by either the first or second artificial intelligence. This starting AI output sentence is referred to here as the starting output sentence. Furthermore, the user input text that triggers the generation of the starting output sentence is called the starting input sentence. Also, of the first and second artificial intelligences, the artificial intelligence to which the starting input sentence is input is called the starting AI.

[0102] The instruction to end an AI-to-AI dialogue is the same as the instruction to start an AI-to-AI dialogue. In other words, while an AI-to-AI dialogue is in progress, the user can input an instruction to end the AI-to-AI dialogue through the information processing system 10 by performing the overlay operation again.

[0103] A distinctive feature of this embodiment is that, even in AI-to-AI dialogue, an AI input sentence with the same structure as the dialogue input sentence D3 described above is generated. Hereinafter, the AI ​​input sentence used in AI-to-AI dialogue will be referred to as the "AI-to-AI input sentence".

[0104] In other words, the starting output sentence, along with a series of utterances between the user and the starting AI immediately preceding the starting output sentence, the speaker information of the series of utterances, and the speaker information of the starting output sentence, is input to the non-starting AI as the first AI-to-AI input sentence. The series of utterances consists of a predetermined number of utterances. The non-starting AI is one of the first and second artificial intelligences that is not the starting AI, i.e., the artificial intelligence that interacts with the starting AI.

[0105] The starting output sentence included in the first AI-to-AI input sentence contains the above-mentioned "[Opinion of XX]" as a special string. Here, "XX" is replaced with the name of the starting AI that output the starting output sentence. Furthermore, the first AI-to-AI input sentence, like the dialogue input sentence D3, includes a series of statements containing the most recent statements between the non-starting AI and the user, as well as speaker information for the series of statements.

[0106] When sending an AI output statement from a non-starting AI that outputs a response statement to a first AI-to-AI input statement to a starting AI for AI-to-AI dialogue, the information processing system 10 inputs a second AI-to-AI input statement to the starting AI, which includes the AI ​​output statement from the non-starting AI, a series of utterances between the user and the non-starting AI up to immediately before the AI ​​output statement, speaker information for the series of utterances, and speaker information for the AI ​​output statement.

[0107] When sending an AI output statement from a starting AI that outputs an AI output statement for a second AI-to-AI input statement to a non-starting AI for AI-to-AI dialogue, the information processing system 10 inputs a third AI-to-AI input statement to the non-starting AI. This third AI-to-AI input statement includes the AI ​​output statement from the starting AI, a series of utterances immediately preceding the AI ​​output statement, speaker information for the series of utterances, and speaker information for the AI ​​output statement. The fourth AI-to-AI input statement is generated in the same manner as the third AI-to-AI input statement.

[0108] According to this embodiment, meaningful information can be obtained from the first artificial intelligence and the second artificial intelligence through such AI-to-AI dialogue.

[0109] In addition, the information processing system 10 of this embodiment has a cascade function that sequentially inputs AI output sentences to multiple artificial intelligences 71A. The cascade function can be used when three or more artificial intelligences 71A are designated as conversation partners from among a pre-prepared group of artificial intelligences 71A dedicated to the cascade function, and when three or more chat screens G1 are displayed on the service provision screen G0.

[0110] The cascading function is achieved by providing a dedicated artificial intelligence 71A for each step of an intelligent task that can be broken down into multiple steps. In other words, the cascading function uses the AI ​​output sentence from the artificial intelligence 71A of each step as input to the artificial intelligence 71A of the next step, thereby realizing the intelligent task using multiple artificial intelligences 71A. By obtaining deliverables based on the intelligent task using multiple artificial intelligences 71A, high-quality deliverables can be obtained efficiently. An example of an intelligent task is the task of writing an article.

[0111] As shown in Figure 10, the process of creating an article can be subdivided into the following steps: deciding on the article's plan, deciding on the article's title, deciding on the article's structure (chapter divisions), creating an outline for each chapter, and creating the main body of each chapter from the outlines.

[0112] When using the cascade function, a dedicated artificial intelligence 71A is assigned to each step based on the user's designation of a conversation partner. Subsequently, the information processing system 10 receives input text from the user to the artificial intelligence 71A located at the very top of the intellectual work process, and inputs the AI ​​output text from the artificial intelligence 71A of each step to the artificial intelligence 71A of the next step.

[0113] In Figure 10, the AI ​​output text of the artificial intelligence 71A located at the starting point of each arrow is input to the artificial intelligence 71A located at the ending point of the arrow. From this, the plan, article title, article structure, summary of each chapter, and the article text of each chapter are created by the corresponding artificial intelligence 71A. The article text of each chapter is created by the artificial intelligence 71A prepared for each chapter. Finally, the article texts of each chapter that have been created are compiled into a single article by the master 71M.

[0114] In Figure 10, the circle symbol drawn in the center of the arrow represents the procedure in which the user selects one of the presented candidates, based on the premise that the AI ​​output sentence from upstream of the arrow presents multiple candidates for the input sentence to the next step.

[0115] This procedure is based on the AI ​​output displayed on the chat screen G1, and the user will use the same chat This is achieved by inputting an AI input message through the input field G15 on the chat screen G1, which indicates the selection of one of several candidates.

[0116] The artificial intelligence 71A takes an AI input sentence that indicates one candidate selected by the user from among several candidates presented to the user as input to the artificial intelligence 71A in the next step. In the first example, the AI ​​input sentence input to the artificial intelligence 71A in the next step is an input sentence that includes information on multiple candidates and information on the candidate selected by the user. In the second example, the AI ​​input sentence is an input sentence that includes only the information on the candidate selected by the user from among multiple candidates.

[0117] When the first example AI input sentence is input to the artificial intelligence 71A in the next step, the artificial intelligence 71A in the next step can generate an AI output sentence by taking into account both the candidate selected by the user and the candidate that was not selected from among multiple candidates. In this case, the artificial intelligence 71A can be instructed in advance to generate an AI output sentence based on the selected candidate, taking into account that the candidate that was not selected by the user was not selected.

[0118] The functions of the information processing system 10 described above are realized by the processor 11 executing processes according to computer programs stored in the storage 15. Next, the details of the processes executed by the processor 11 will be explained using Figures 11 to 15.

[0119] When the information terminal 60 requests access to the information processing system 10, the processor 11, after user authentication, starts the process shown in Figure 11 and communicates with the information terminal 60 to display the service provision screen G0 on the display device 61A of the information terminal 60 (S110).

[0120] Subsequently, the processor 11 waits until an operation is performed on the service provision screen G0 via the input device 61B (S120), and once an operation is performed (Yes in S120), it identifies the operation that was performed (S130).

[0121] If the operation identified in S130 is a text input operation to input fields G15 and G35 (Yes in S140), the processor 11 accepts the text input operation to input fields G15 and G35 and displays the input text in the corresponding input fields G15 and G35 (S145). After that, the processor 11 returns to the process of S120.

[0122] If the operation identified in S130 is the pressing of the send buttons G18 and G371 (Yes in S150), the processor 11 determines that a text transmission instruction has been input and, as shown in Figure 4A, sends the text entered in the input fields G15 and G35 of the same chat screen G1 and G3 as the AI ​​input text to the artificial intelligence 71 associated with the same chat screen G1 and G3 as the pressed send buttons G18 and G371 (S160).

[0123] The processor 11 further displays the AI ​​input sentence sent in S160, and the AI ​​output sentence received from the artificial intelligence 71 that received the AI ​​input sentence as a response to this AI input sentence, in the form of speech bubbles on the chat display screens G13 and G33 of the chat screens G1 and G3 associated with the artificial intelligence 71 that received the AI ​​input sentence (S165). After that, the processor 11 returns to the processing of S120.

[0124] In S160, if no text is entered in the corresponding input fields G15 and G35, the processor 11 does not send an AI input sentence to the artificial intelligence 71, nor does it execute the process in S165, and returns to the process in S120.

[0125] If the operation identified in S130 is the press of the simultaneous transmission button G19 (S170 If the answer is Yes, the processor 11 determines that a message to send text to multiple users has been entered and, as shown in Figure 4B, sends the text entered in the input field G15 as an AI input message to all of the multiple artificial intelligences 71A associated with each of the multiple chat screens G1 displayed on the service provision screen G0 (S180). The text to be sent to multiple users is, for example, the text entered in the input field G15 of the same chat screen G1 as the message sent button G19 that was pressed.

[0126] The processor 11 further displays the AI ​​input sentence sent in S180, and the AI ​​output sentences received from each artificial intelligence 71A that received the AI ​​input sentence, in the form of speech bubbles on the chat display screen G13 of the chat screen G1 associated with each artificial intelligence 71A (S185). After that, the processor 11 returns to the processing of S120.

[0127] If the operation identified in S130 is a transfer operation (Yes in S190), the processor 11 determines that a transfer instruction has been entered and displays a processed transfer text D1 in the input field G15 where the transfer text was dropped, with the speaker's name appended to the end of the transfer text (i.e., the transfer text) as a special string "[Opinion of XX]" (S200).

[0128] Subsequently, when the send button G18 on the same chat screen G1 as the input field G15 is pressed, the processor 11 sends the processed transfer text D1, the series of utterances D2, related text D4, and the dialogue input text D3, which includes the speaker information for these, to the artificial intelligence 71A associated with the same chat screen G1 as the pressed send button G18 (S205).

[0129] As described above, and further as shown in Figure 7, the series of statements D2 transmitted in S205 is a series of statements D2 that shows the content of the dialogue between the artificial intelligence 71A (i.e., the first artificial intelligence) and the user, which is associated with the chat screen G1 from which the drag-and-drop statement (i.e., the forwarded statement) was dragged, and is a group of predetermined past statements adjacent to the forwarded statement. The series of statements D2 explains the context of the dialogue leading up to the forwarded statement.

[0130] As described above, related text D4 is a series of utterances that represent the content of the conversation between the AI ​​71A (i.e., the second AI) associated with the drop destination chat screen G1 and the user, and is a group of predetermined utterances from the most recent to the past. This related text D4 is sent to the second AI in order to generate an AI output sentence that takes into account the past conversations between the second AI and the user.

[0131] Subsequently, the processor 11 displays the processed transfer sentence D1 contained in the dialogue input sentence D3 sent in S205, and the AI ​​output sentence received from the artificial intelligence 71A (i.e., the second artificial intelligence) to which the dialogue input sentence D3 was sent as a response to the dialogue input sentence D3, in the form of a speech bubble on the chat display screen G13 of the chat screen G1 associated with this artificial intelligence 71A (i.e., the second artificial intelligence) (S209). After that, the processor 11 returns to the processing of S120.

[0132] If the operation identified in S130 is a superposition operation (Yes in S210), the processor 11 determines whether the two artificial intelligences 71A corresponding to the two chat screens G1 that were the subject of the superposition operation are currently in an AI-to-AI dialogue with each other (S220).

[0133] If it is determined that there is no AI-to-AI dialogue in progress (No in S220), the processor 11 recognizes that the superposition operation is an instruction to start an AI-to-AI dialogue and begins processing the AI-to-AI dialogue for these two artificial intelligences 71A (S230). After that, the processor 11 returns to the processing in S120.

[0134] On the other hand, if it is determined in S220 that an AI-to-AI dialogue is in progress (Yes in S220), the processor 11 determines that the superposition operation is an instruction to terminate the AI-to-AI dialogue and terminates the AI-to-AI dialogue processing for the two AI-to-AI dialogues 71 (S240). After that, the processor 11 returns to the processing in S120.

[0135] In S230, the processor 11 can execute the AI-to-AI dialogue process shown in Figure 13. When the AI-to-AI dialogue process shown in Figure 13 is started, the processor 11 accepts text input from the input field G15 of the chat screen G1 of one of the two artificial intelligences 71A that are the target of the AI-to-AI dialogue (S231).

[0136] Subsequently, when the send button G18 on the chat screen G1 is pressed, the processor 11 sends the entered text as the starting input text to the starting AI (S232). The starting AI is the artificial intelligence 71A that received the text input from the input field G15, which is one of the two artificial intelligences 71A that were the target of the AI-to-AI dialogue.

[0137] After transmission in S232, the processor 11 displays the starting input sentence transmitted in S232, and the AI ​​output sentence (i.e., starting output sentence) received from the artificial intelligence 71A (i.e., the starting AI) that received the starting input sentence as a response to the starting input sentence, in the form of a speech bubble on the chat display screen G13 of the chat screen G1 associated with the artificial intelligence 71A (S233).

[0138] Furthermore, the processor 11 transmits the first AI-to-AI input sentence, which includes the starting output sentence, to the non-starting AI (S234). After transmission in S234, the processor 11 displays the AI ​​output sentence received from the artificial intelligence 71A (i.e., the non-starting AI) to which the first AI-to-AI input sentence was sent, as a response to the starting output sentence and the first AI-to-AI input sentence transmitted in S234, in the form of a speech bubble on the chat display screen G13 of the chat screen G1 associated with this artificial intelligence 71A (S235).

[0139] Furthermore, processor 11 sends a second inter-AI input sentence, which includes the AI ​​output sentence received in S235, to the starting AI (S236).

[0140] After transmission in S236, the processor 11 implements AI-to-AI dialogue by repeatedly performing the same processing as in S233, S234, and S235 related to the transmitted AI-to-AI input statement until an instruction to terminate the AI-to-AI dialogue is received. When understanding the processing similar to S233, the starting input statement should be read as the AI ​​output statement from a non-starting AI included in the AI-to-AI input statement received immediately before.

[0141] In addition, if the operation identified in S130 is not a text input operation in input fields G15 or G35, a press of the send button G18, a press of the mass send button G19, a transfer operation, or an overlay operation (No in S140, No in S150, No in S170, No in S190, No in S210), and also not an operation to close the service provision screen G0 (No in S260), then the processor 11 executes the processing corresponding to the other identified operation (S270).

[0142] Other examples of operations include pressing the call button G373, the notepad button G375, the output button G379, and operations related to rearranging the chat screen G1. The processor 11 can control the display of the display device 61A to change the left-right arrangement of multiple chat screens G1 displayed on the service provision screen G0 according to the rearrangement operation of the chat screen G1 (S270). After executing the process in S270, the processor 11 returns the process to S120. The processor 11 then returns the service provision screen When the operation to close is performed (Yes in S260), the operation acceptance process shown in Figures 11 and 12 is terminated.

[0143] Next, using Figure 14, the contents of the cascade-related processing performed by the processor 11 to realize the cascade function will be explained. When the cascade-related processing shown in Figure 14 is started, the processor 11 receives an input sentence from the user to the artificial intelligence 71A, which is located at the very top of the intelligent work realized by the cascade function, that is, the artificial intelligence 71A that realizes the first step of the intelligent work (S310).

[0144] Specifically, the processor 11 accepts text input from the user through the input field G15 of the chat screen G1, which is associated with the top-level artificial intelligence 71A, i.e., the first-step artificial intelligence 71A (S310).

[0145] When text input is received, the processor 11 sends the input text to the artificial intelligence 71A in the first step as an AI input sentence (S320).

[0146] Subsequently, the processor 11 displays the AI ​​input sentence and the AI ​​output sentence received from the artificial intelligence 71A of the first step as a response to the AI ​​input sentence in the form of a speech bubble on the chat display screen G13 of the chat screen G1 of the artificial intelligence 71 of the first step (S330).

[0147] The AI ​​output sentence from artificial intelligence 71A in the first step presents multiple candidates for the input sentence to artificial intelligence 71A in the next step (i.e., the second step).

[0148] After processing in S330, the processor 11 accepts the user's selection from among the multiple candidates included in the AI ​​output sentence received in S330 via the input field G15 (S340). Specifically, it accepts the input operation via the input field G15 of text indicating the candidate to be selected by the user as the selection content (S340). Once the text indicating the selection content is entered, the processor 11 transmits the corresponding selection content to the artificial intelligence 71A in the first step (S350).

[0149] After processing in S350, the processor 11 receives the AI ​​output sentence output from the first step artificial intelligence 71A and sends the AI ​​output sentence to the second step artificial intelligence 71A as an AI input sentence (S360). The second step artificial intelligence 71A is the artificial intelligence 71A that realizes the second step of the intelligent work.

[0150] The AI ​​output sentence produced by the artificial intelligence 71A in the first step, or in other words, the AI ​​input sentence to the artificial intelligence 71A in the second step, is the AI ​​input sentence relating to the first or second example described above with respect to the cascade function.

[0151] In other words, according to the first example, the AI ​​input sentence is an input sentence that includes information on the multiple candidates mentioned above and information on the candidate selected by the user. According to the second example, the AI ​​input sentence is an input sentence that includes only the information on the candidate selected by the user from among the multiple candidates.

[0152] After processing in S360, processor 11 receives the AI ​​output sentence from the artificial intelligence 71 of the second step as a response sentence to this AI input sentence. Processor 11 displays the received AI output sentence in the form of a speech bubble on the chat display screen G13 of the chat screen G1 associated with the artificial intelligence 71 of the second step (S370). The AI ​​output sentence from the artificial intelligence 71A of the second step presents multiple candidates for the input sentence to the artificial intelligence 71A of the next step (i.e., the third step).

[0153] In the subsequent S380, the processor 11 receives the user's selection from multiple candidates for the input sentence to the artificial intelligence 71A in the next step, which is included in the AI ​​output sentence received in S370, via the input field G15. When text representing the selection is entered into the input field G15, the processor 11 transmits the corresponding selection to the artificial intelligence 71A in the second step (S390).

[0154] After processing in S390, the processor 11 receives the AI ​​output sentence output from the artificial intelligence 71A of the second step and sends the AI ​​output sentence to the artificial intelligence 71A of the third step as a new AI input sentence (S400).

[0155] Subsequently, the processor 11 receives the AI ​​output sentence from the artificial intelligence 71 of the third step and repeatedly performs the same processing as in S370, S380, S390, and S400, updating the steps of the intelligent work.

[0156] However, if the AI ​​output sentence received in S370 does not contain any candidates, the processor 11 may not accept input for selection in S380 and may send the AI ​​output sentence received in S370 to the artificial intelligence 71 in the next step (S400).

[0157] The cascading function is realized by the processing described above, which is performed by processor 11. In addition, according to the information processing system 10 of this embodiment, the processor 11 repeatedly executes the master-related processing shown in Figure 15 to provide the master 71M with the content of the dialogue between each artificial intelligence 71A and the user.

[0158] Specifically, when a new message appears in one of the multiple chat screens G1 (Yes in S410), the processor 11 adds this message to the chat display screen G33 of the chat screen G3 in the form of a speech bubble (S420). If the message is from the artificial intelligence 71A, the processor 11 modifies the message by adding "[○○'s opinion]" to the end, and then adds the modified message to the chat display screen G33. The name of the speaker is inserted in place of "○○". The processor 11 then sends the message (modified message) to the master 71M along with instructions to record it as a log (S430). At this time, inventor information is provided to the master 71M.

[0159] The information processing system 10 of this embodiment, as described above, is configured to acquire generated sentences (i.e., AI output sentences) from each of the multiple artificial intelligences 71. The information processing system 10 is configured to acquire input sentences from the user through the input device 61B and to transmit the input sentences as AI input sentences to at least one of the multiple artificial intelligences 71.

[0160] The information processing system 10 is configured to display the AI ​​output sentences obtained from each of the multiple artificial intelligences 71 on the display device 61A via the service provision screen G0.

[0161] When the information processing system 10 receives a transmission instruction from the user via the input device 61B by pressing the transmit button 18G (Yes in S150), it transmits the input text to one of the multiple artificial intelligences 71 designated by the user, i.e., the artificial intelligence 71 corresponding to the pressed transmit button 18G (S160).

[0162] When the information processing system 10 receives a mass transmission instruction from the user via the input device 61B by pressing the mass transmission button 19G (Yes in S170), it transmits the input message to all of the artificial intelligences 71A except the master 71M (S180).

[0163] With the information processing system 10 configured in this way, a user can simultaneously send a common input message to multiple artificial intelligences 71A. Therefore, according to this embodiment, it is possible to provide a highly convenient information processing system 10 when a user uses multiple artificial intelligences 71.

[0164] According to this embodiment, when the information processing system 10 receives a transfer instruction via the input device 61B (Yes in S190), it transmits the content of the dialogue between the user and the first artificial intelligence (the source artificial intelligence) among the multiple artificial intelligences 71A, through AI input sentences and AI output sentences, to a second artificial intelligence (the destination artificial intelligence) among the multiple artificial intelligences 71A, which is different from the first artificial intelligence (S205). The information processing system 10 obtains the AI ​​output sentence from this second artificial intelligence and displays it on the display device 61A (S209).

[0165] The transfer operation is a drag-and-drop operation of the spoken text, and the first artificial intelligence and the second artificial intelligence correspond to the artificial intelligence from which the text is dragged and the artificial intelligence to which it is dragged, respectively. In other words, according to this embodiment, the user can specify the first artificial intelligence and the second artificial intelligence by dragging and dropping.

[0166] With the information processing system 10 configured in this way, the user can obtain a response from a second artificial intelligence to the content of their interaction with the first artificial intelligence. For example, the user can obtain a different idea from the second artificial intelligence that takes into account the thoughts of the first artificial intelligence.

[0167] For example, regarding a certain theme, it is possible to obtain a woman's perspective from a second artificial intelligence that incorporates the man's perspective obtained from a first artificial intelligence. The examples of virtual personality combinations for the combination of the first and second artificial intelligences here are not limited to combinations of men and women, but include various combinations of people, such as combinations of people from different residential areas, combinations of people from different generations, and combinations of people with different areas of expertise. Therefore, according to this embodiment, it is possible to meaningfully utilize multiple artificial intelligences 71, each with a different virtual personality, in other words, each with a different role.

[0168] Furthermore, the dialogue input sentence D3 transmitted to the second artificial intelligence through the transfer operation includes the content of the conversation between the user and the first artificial intelligence leading up to the utterance of the transferred sentence. However, the dialogue content included in the dialogue input sentence D3 is not the entirety of the conversation between the user and the first artificial intelligence, but rather a part related to the transferred sentence. By selectively providing the second artificial intelligence with only a part of the dialogue content, rather than the whole, it is possible to obtain an appropriate response from the second artificial intelligence to the statement that the user wants to draw attention to.

[0169] The forwarded text is, in most cases, the most recent statement from the group of statements generated during the interaction between the user and the first artificial intelligence. In this case, the dialogue input text D3 can store the most recent part of the dialogue content exchanged between the user and the first artificial intelligence. By providing the most recent part of the dialogue content to the second artificial intelligence, the user can obtain the second artificial intelligence's response to the most recent dialogue between the user and the first artificial intelligence.

[0170] In addition, in this embodiment, when the information processing system 10 receives an instruction from the user to start an AI-to-AI dialogue via the input device 61B through a superposition operation (Yes in S210), it realizes a dialogue between the first artificial intelligence and the second artificial intelligence, which are two artificial intelligences 71A corresponding to the two chat screens G1 that were the subject of the superposition operation. The first artificial intelligence and the second artificial intelligence here correspond to the artificial intelligences 71A specified by the user through the superposition operation.

[0171] Specifically, the information processing system 10 realizes dialogue between the first artificial intelligence and the second artificial intelligence by executing the AI-to-AI dialogue process shown in Figure 13, by sending the AI ​​output sentence obtained from the first artificial intelligence as an AI input sentence to a second artificial intelligence different from the first artificial intelligence (S234), and by sending the AI ​​output sentence obtained from the second artificial intelligence to the first artificial intelligence as an AI input sentence (S236).

[0172] With the information processing system 10 configured in this way, users can view the results of discussions between artificial intelligences regarding a theme or topic of interest. Such discussion results can, for example, be useful for the user's deeper thinking on that theme or topic.

[0173] In addition, in this embodiment, with respect to the cascade function, the information processing system 10, starting with a user instruction to send an AI input sentence to the first step artificial intelligence 71A, sends the AI ​​output sentence of the first step artificial intelligence 71A as an AI input sentence to the second step artificial intelligence 71A (S360), and sends the AI ​​output sentence of the second step artificial intelligence 71A as an AI input sentence to the third step artificial intelligence 71A (S400).

[0174] According to this embodiment, each step of an intellectual task can be realized using different artificial intelligences 71A. Compared to performing an intellectual task such as writing an article through interaction using a single artificial intelligence 71A, this method allows for the creation of a product that is the result of the combined thoughts of multiple artificial intelligences 71A, or a product based on multifaceted thinking, thereby enabling the efficient acquisition of high-quality products.

[0175] In addition, according to this embodiment, the information processing system 10 transmits the content of the conversation between the user and the multiple artificial intelligences 71A to a master 71M, which is a specific artificial intelligence 71M different from the multiple artificial intelligences 71A (S430). When the information processing system 10 receives a transmission instruction to the master 71M by pressing the transmit button G371, it transmits the text entered in the input field G35 to the master 71M as an AI input sentence (S160).

[0176] Therefore, the user can send instructions to the master 71M as AI input sentences to perform processing based on the content of the user's dialogue with multiple artificial intelligences 71A, and obtain the response as an AI output sentence (S165). The response from the master 71M (AI output sentence) is displayed on the display device 61A through the chat display screen G33 of the chat screen G3 (S165).

[0177] With the information processing system 10 configured in this way, the user can obtain a response statement from the master 71M that comprehensively considers the interactions between the user and multiple artificial intelligences 71A. For example, the user can have the master 71M generate a summary or minutes of the conversations between the user and multiple artificial intelligences 71.

[0178] In addition, according to this embodiment, the multiple artificial intelligences 71A include two or more artificial intelligences 71 that play different roles. Therefore, the user can receive meaningful information from the multiple artificial intelligences 71A.

[0179] Furthermore, the multiple artificial intelligences 71A may include an artificial intelligence that acts as a "devil's advocate," and the information processing system 10 may be configured so that the user can specify the artificial intelligence 71A that acts as a "devil's advocate" as their conversation partner.

[0180] The AI ​​71A that plays the role of "the devil's advocate" is an AI 71A that is programmed to always speak in a critical or counter-critical position to statements made by the user or other AI 71A. By providing such an AI 71A, the user can have "the devil's advocate" as a conversation partner. By designating AI 71A to act as the "spokesperson for evil," it becomes possible to gain a deeper understanding of various themes through dialogue with AI 71A.

[0181] In addition, in this embodiment, the service provision screen G0 displays a chat screen G1 for each artificial intelligence 71A. As a result, the display screen of the display device 61A is partitioned. That is, the content of the conversation between the user and one corresponding artificial intelligence 71A is displayed in each of the multiple areas corresponding to the multiple chat screens G1. In other words, the information processing system 10 controls the display device 61A via the service provision screen G0 to display the content of the conversation between the user and one corresponding artificial intelligence 71A from among the multiple artificial intelligences 71A.

[0182] Therefore, according to this embodiment, the content of the dialogue between multiple artificial intelligences 71A and the user can be displayed to the user in an easy-to-understand manner.

[0183] According to this embodiment, each chat screen G1, G3 is configured as a movable display window. That is, the information processing system 10 is configured to accept rearrangement operations from the user for multiple chat screens G1, G3 via the input device 61B, and to change the order of the multiple chat screens G1, G3 on the service provision screen G0 according to the rearrangement operation. Therefore, according to this embodiment, the information processing system 10 can provide the user with a display screen that is easier for the user to understand regarding the content of the conversation between the user and multiple artificial intelligences 71.

[0184] According to this embodiment, the chat display screen G13 of each chat screen G1 displays the content of the conversation between the user and a corresponding artificial intelligence 71A by arranging the user's input text and the AI ​​output text from the corresponding artificial intelligence 71A in chronological order. Each chat screen G1 also displays an input field G15 for writing input text to the corresponding artificial intelligence 71A.

[0185] The information processing system 10 is configured to determine that a transfer instruction has been entered when a utterance (including an AI output) displayed as dialogue content on the chat display screen G13 is dragged and dropped into the input field G15, and then transmits the dragged utterance to the artificial intelligence 71A (the AI ​​at the drop destination) corresponding to the input field G15 where the utterance was dropped.

[0186] When a dragged speech statement is sent to the AI ​​71A at the drop location, the dragged speech statement (AI output) is sent to the AI ​​71A along with a portion of the original dialogue that explains the context in which the dragged speech statement (AI output) was generated.

[0187] With the information processing system 10 configured in this way, user operability regarding transfer operations can be improved, and the user can obtain an appropriate response message from the drop destination artificial intelligence 71A with simple operations.

[0188] In addition, according to this embodiment, the instruction to start an AI-to-AI dialogue can be realized by overlaying the chat screen G1. Therefore, the user experience regarding inputting the instruction to start an AI-to-AI dialogue is excellent.

[0189] [Second Embodiment] Next, the information processing system 10 of the second embodiment will be described. The information processing system 10 of the second embodiment is an addition of new functions to the information processing system 10 of the first embodiment. The hardware configuration of the information processing system 10 of the second embodiment is the same as that of the first embodiment. be.

[0190] Therefore, in the following description of the second embodiment, we will selectively describe the configurations related to the functions added to the information processing system 10 of the second embodiment. Components in the information processing system 10 of the second embodiment that are the same as those in the first embodiment are denoted by the same reference numerals and their descriptions will be omitted.

[0191] In this embodiment, one or more artificial intelligences 71B having a scoring function are provided as at least one of the plurality of artificial intelligences 71A described above. The artificial intelligence 71B is an artificial intelligence 71A having a scoring function.

[0192] The artificial intelligence 71B is configured to score an AI input sentence when instructed to do so by the user, and to generate and output an AI output sentence that includes the scoring result. In other words, the artificial intelligence 71B is configured to generate and output an AI output sentence that includes the evaluation score for the AI ​​input sentence.

[0193] The information processing system 10 can prepare one or more artificial intelligences 71B on the AI ​​server 70 by inputting instructions to create artificial intelligences 71B to the AI ​​server 70. As shown in Figure 16, the information processing system 10 can prepare one or more artificial intelligences 71B on the AI ​​server 70 by inputting instructions to the AI ​​server 70, along with information about the virtual personality, which include instructions to "score the AI ​​input sentence when instructed by the user to score / evaluate it, and generate and output an AI output sentence that includes the scoring result."

[0194] In this embodiment, the information processing system 10 does not specify explicit scoring criteria to the AI ​​server 70 when instructing it to generate artificial intelligence 71B. Each artificial intelligence 71B operates to score the AI ​​input sentence based on its own ideas without specifying scoring criteria. Each artificial intelligence 71B is assigned an individual virtual personality. The scoring of the artificial intelligence 71B is influenced by the virtual personality. That is, different scoring criteria are applied to the AI ​​input sentence between artificial intelligence 71Bs assigned different virtual personalities.

[0195] Figure 17 illustrates the content of the conversation between the user and the artificial intelligence 71B as displayed on the chat display screen G13. According to this embodiment, the user can input the message "Please evaluate the idea I am about to explain" to the artificial intelligence 71B as an AI input sentence through the input field G15. A concrete example of an idea is an concept. Subsequently, the user can input an AI input sentence explaining the idea to the artificial intelligence 71B through the input field G15.

[0196] In response to this AI input, artificial intelligence 71B outputs a response sentence as AI output sentence, which includes an evaluation score for the user's idea and details of the evaluation. Through this scoring function, the user can obtain an objective evaluation of their own ideas and thoughts through dialogue with artificial intelligence 71B.

[0197] Furthermore, the scoring function can be used not only in the dialogue between the user and artificial intelligence 71B, but also in the dialogue between artificial intelligence 71A and artificial intelligence 71B. According to this embodiment, the processor 11, as in the first embodiment (see Figure 12), starts AI-to-AI dialogue processing (Figure 13) (S230) when an overlay operation is performed (Yes in S210).

[0198] When the user provides the artificial intelligence 71A with a dialogue theme, such as brainstorming ideas, through text input to the artificial intelligence 71A (S231), the artificial intelligence 71A's response to the theme is generated as an AI output sentence and displayed on the screen (S232, S233), and this AI output sentence is input to the artificial intelligence 71B (S234).

[0199] Artificial intelligence 71B scores the response of artificial intelligence 71A included in this AI output sentence, and the AI ​​output sentence, including the score and details of the evaluation, is displayed on the screen and input into artificial intelligence 71A (S235, S236).

[0200] Based on the evaluation score, artificial intelligence 71A generates an AI output sentence containing a new answer to the above theme, and this AI output sentence is input to artificial intelligence 71B. In this way, the dialogue between artificial intelligence 71A and 71B is repeated, with artificial intelligence 71A's statements being scored by artificial intelligence 71B.

[0201] While the AI-to-AI dialogue process is running, the processor 11 concurrently executes the termination control process shown in Figure 18. In the termination control process, each time the AI ​​output sentence of the artificial intelligence 71A is scored (Yes in S510), the processor 11 determines whether the evaluation score by the artificial intelligence 71B is above a threshold (S520). Here, a higher evaluation is considered to correspond to a higher evaluation score.

[0202] If the evaluation score is below the threshold (No in S520), processor 11 decides to continue the AI-to-AI dialogue process (S530) and returns to the process in S510. On the other hand, if the score of artificial intelligence 71B is above the threshold (Yes in S520), processor 11 decides to terminate the AI-to-AI dialogue process (S540). Processor 11 terminates the AI-to-AI dialogue process (Figure 13) when the display of the evaluation score on the screen is complete. After that, processor 11 terminates the termination control process.

[0203] In this embodiment, the processor 11 repeatedly causes the artificial intelligences 71A and 71B to perform a dialogue until a response of predetermined quality is obtained on a theme specified by the user. When a response of predetermined quality is obtained, the processor 11 terminates the dialogue between the artificial intelligences 71A and 71B without any instruction from the user.

[0204] According to this embodiment, by utilizing the scoring function described above, the user can obtain an appropriate answer on a desired topic from the artificial intelligence 71A. Therefore, the user can conveniently use the dialogue function between the artificial intelligences 71A and 71B.

[0205] [Third Embodiment] Next, the information processing system 10 of the third embodiment will be described. The information processing system 10 of the third embodiment is the information processing system 10 of the first embodiment with new functions added. The hardware configuration of the information processing system 10 of the third embodiment is the same as that of the first embodiment.

[0206] Therefore, in the following description of the third embodiment, we will selectively describe the configuration of functions added to the information processing system 10 of the first embodiment. Components in the information processing system 10 of the third embodiment that are the same as those in the first embodiment will be denoted by the same reference numerals and their descriptions will be omitted. The additional functions described below may be provided in the information processing system 10 of the second embodiment instead of the first embodiment.

[0207] The information processing system 10 of this embodiment has an automatic relay function as one of the cascade functions described in the first embodiment. In the automatic relay function, as in the first embodiment, the AI ​​output sentence from the AI ​​71A of the upstream step is input to the AI ​​71A of the adjacent next step downstream, without presenting the user with candidate input sentences for the next step.

[0208] In other words, with the automatic relay function, the AI ​​output sentences from each artificial intelligence 71A, from the first step's artificial intelligence 71A to the final step's artificial intelligence 71A, are input to the downstream, adjacent next step's artificial intelligence 71A without any user selection of candidates. This allows the user to have multiple artificial intelligences 71A automatically execute an intelligent task that can be broken down into multiple steps, without any user selection of candidates.

[0209] Examples of intellectual work include writing articles, as mentioned above, and examples of deliverables include articles. Another example of intellectual work is MVVSS, a framework that systematically organizes the five elements that constitute the foundation of an organization. MVVSS consists of five elements: Mission (M), Vision (V), Values ​​(V), Spirit (S), and Slogan (S). An automated relay function can be used in formulating this MVVSS.

[0210] According to this embodiment, the automatic relay function does not have a function to allow the user to select a candidate, but is configured to accept additional instructions for the artificial intelligence 71A. The user can perform an additional operation to instruct artificial intelligence 71A_2 in the transmission path of AI output sentences between any two adjacent artificial intelligences 71A_1, 71A_2 (see Figure 19) from the multiple artificial intelligences 71A from the first step to the final step.

[0211] According to this embodiment, the information processing system 10 displays the chat screen G1 as a single virtual card and is configured to perform an action of flipping the card over according to user input, as shown in Figure 20. For example, the user can flip the card over by clicking the upper left area of ​​the chat screen G1.

[0212] The back of the card is provided with an input field G16 for the user to input instructions for the corresponding artificial intelligence 71A (hereinafter referred to as "additional instructions"). By inputting additional instructions for the artificial intelligence 71A into the input field G16, the user can indirectly control the AI ​​output sentences generated by the artificial intelligence 71A.

[0213] For example, when creating an MVVSS using the automatic relay function, the user can input additional instructions to limit the direction of the MVVSS in input field G16, thereby causing the artificial intelligence 71A to generate an MVVSS in the direction desired by the user. In other words, it is possible to prevent the artificial intelligence 71A from generating an MVVSS in a direction not desired by the user.

[0214] When an execution instruction for the automatic relay function is input to the service provision screen G0, with two or more artificial intelligence 71A units used in the automatic relay function, from the first step to the final step, the processor 11 starts the relay-related processing shown in Figure 21. When the relay-related processing starts, the processor 11 sets the variable M to the value "1" (S610).

[0215] The processor 11 further accepts input operations for additional instructions to the artificial intelligence 71A from the first step to the final step via the service provision screen G0 (S620). The user can optionally input additional instructions to one or more of the artificial intelligence 71A from the first step to the final step via the input field G16.

[0216] Subsequently, the processor 11, similar to S310, accepts text input from the user via the input field G15 for the artificial intelligence 71A to realize the first step of intelligent work (S630).

[0217] When text input is received, processor 11 determines whether additional instructions have been given to the artificial intelligence 71A in step M (S640). If the determination is made (No in S640), the processor 11 inputs the entered text as an AI input sentence to the artificial intelligence 71A in the M step (S650).

[0218] If the processor determines that additional instructions have been entered (Yes in S640), it inputs the input text, with the additional instructions attached, to the artificial intelligence 71A in the M step as an AI input sentence (S655). As a result, the processor causes the artificial intelligence 71A to generate an AI output sentence with the additional instructions given to it in addition to the instructions given to it when it was generated in the M step.

[0219] Subsequently, the processor 11 displays the AI ​​input sentence to the artificial intelligence 71A in the M step, and the AI ​​output sentence received from the artificial intelligence 71A in the M step as a response to the AI ​​input sentence, in the form of a speech bubble on the chat display screen G13 of the chat screen G1 of the artificial intelligence 71A in the M step (S660).

[0220] The processor 11 further increments the variable M by 1, that is, updates the variable M to a value that has been increased by 1 (S670). Then, before inputting the AI ​​output sentence received immediately from the artificial intelligence 71A of the (M-1) step to the artificial intelligence 71A of the next step, i.e., the artificial intelligence 71A of the M step, the processor 11 determines whether any additional instructions have been input to the artificial intelligence 71A of the M step (S680).

[0221] If it determines that no additional instructions have been entered (No in S680), the processor 11 inputs the AI ​​output sentence received from the artificial intelligence 71A in the (M-1) step as the AI ​​input sentence to the artificial intelligence 71A in the M step (S690).

[0222] If the processor determines that an additional instruction has been input (Yes in S680), the processor 11 inputs the AI ​​output sentence received from the artificial intelligence 71A in the (M-1) step, with the additional instruction attached, to the artificial intelligence 71A in the M step as an AI input sentence (S695).

[0223] Subsequently, the processor 11 displays the AI ​​input sentence and the AI ​​output sentence received from the artificial intelligence 71A in the Mth step as a response to the AI ​​input sentence in the form of speech bubbles on the chat display screen G13 of the chat screen G1 of the artificial intelligence 71A in the Mth step (S700).

[0224] After processing in S700, the processor 11 determines whether the Mth step is the final step (S710). If the processor 11 determines that the Mth step is not the final step (No in S710), it repeatedly executes the processes in S670 to S710, thereby inputting the AI ​​output sentence from the artificial intelligence 71A of each step to the artificial intelligence 71A of the next step. If it determines that it is the final step (Yes in S710), the processor 11 terminates the automatic relay-related processing.

[0225] According to this embodiment, the user can have multiple artificial intelligences 71A automatically execute an intelligent task that can be broken down into multiple steps using the automatic relay function, and obtain the output related to that intelligent task.

[0226] The above describes an example of utilizing additional instructions with the automatic relay function, but the use of additional instructions is not limited to the automatic relay function. For example, the information processing system 10 may be configured to accept additional instructions for each artificial intelligence 71A through the service provision screen G0, not limited to the automatic relay function. Acceptance can be achieved using the method shown in Figure 20. When an additional instruction is set, the processor 11 can attach the additional instruction when sending an AI input sentence to the corresponding artificial intelligence 71A. With such additional instruction settings, the user can, The behavior of each artificial intelligence 71A can be indirectly controlled.

[0227] [Other embodiments] This disclosure is not limited to the embodiments described above, and various other embodiments are possible. For example, in the above embodiment, the information processing system 10 provides an interactive service as a web service, but a computer program that realizes the functions of the information processing system 10 may be installed on the information terminal 60, and the information terminal 60 may be configured to communicate with the AI ​​server 70 without going through the information processing system 10.

[0228] The function of one component in the above embodiment may be distributed among multiple components. The functions of multiple components may be integrated into one component. Some of the components of the above embodiment may be omitted. Any aspect of the technical concept specified by the claims constitutes an embodiment of the present disclosure.

[0229] [Technical concepts disclosed in this specification] This specification can be understood to disclose the following technical concepts: [Item 1] A generated text acquisition unit configured to acquire generated text from each of multiple artificial intelligences, A display control unit configured to display the generated sentences obtained from each of the aforementioned multiple artificial intelligences on a display device, A transmission unit configured to acquire input text from a user through an input device and transmit the input text to at least one of the plurality of artificial intelligences, Equipped with, The generated sentence includes a response sentence to the input sentence and is generated by the corresponding artificial intelligence. The aforementioned transmitting unit When a first instruction is input from the user through the input device, the input sentence is transmitted to one of the multiple artificial intelligences designated by the user. If a second instruction is input from the user through the input device, the input sentence is transmitted to the plurality of artificial intelligences. An information processing system configured in such a way. [Item 2] The transmission unit is configured to transmit, when a third instruction is input from the user through the input device, the content of the dialogue between the user and the first artificial intelligence among the plurality of artificial intelligences, through the input sentence and the generated sentence, to a second artificial intelligence among the plurality of artificial intelligences, which is different from the first artificial intelligence. The generated text acquisition unit is configured to acquire, as the generated text, a response text to the dialogue content from the second artificial intelligence. The information processing system described in item 1. [Item 3] The first artificial intelligence and the second artificial intelligence are each the artificial intelligence specified by the user from among the plurality of artificial intelligences. The information processing system described in item 2. [Item 4] The transmitting unit is configured to selectively transmit a portion of the dialogue between the user and the first artificial intelligence to the second artificial intelligence when the third instruction is input. The information processing system described in item 2 or item 3. [Item 5] A portion of the aforementioned dialogue content includes the latest input sentence and generated sentence exchanged between the user and the first artificial intelligence. The information processing system described in item 4. [Item 6] The transmission unit is configured to cause the first and second artificial intelligences to interact when a fourth instruction is input from the user through the input device. This is done by transmitting the generated sentence obtained from the first artificial intelligence among the plurality of artificial intelligences to a second artificial intelligence, which is different from the first artificial intelligence, and by transmitting the generated sentence obtained from the second artificial intelligence to the first artificial intelligence. The information processing system described in item 1. [Item 7] The first artificial intelligence and the second artificial intelligence are each the artificial intelligence specified by the user from among the plurality of artificial intelligences. The information processing system described in item 6. [Item 8] The transmission unit is configured such that, when a fifth instruction is input from the user through the input device, it transmits the generated sentence obtained from the first artificial intelligence among the plurality of artificial intelligences to a second artificial intelligence, which is different from the first artificial intelligence, and transmits the generated sentence obtained from the second artificial intelligence to a third artificial intelligence, which is different from both the first and second artificial intelligences. The information processing system described in item 1. [Item 9] The aforementioned transmitting unit The content of the conversation between the user and each of the aforementioned multiple artificial intelligences is transmitted to a specific artificial intelligence that is different from the aforementioned multiple artificial intelligences. If a sixth instruction is input from the user through the input device, the input sentence is transmitted to the specific artificial intelligence. It is configured in such a way, The generated sentence acquisition unit acquires the generated sentence based on the input sentence transmitted to the specific artificial intelligence from the specific artificial intelligence, The display control unit is configured to display the generated sentence obtained from the specific artificial intelligence on the display device. An information processing system described in any one of items 1 through 8. [Item 10] The aforementioned multiple artificial intelligences include two or more artificial intelligences that play different roles. An information processing system described in any one of items 1 through 9. [Item 11] The aforementioned two or more artificial intelligences include an artificial intelligence that acts as a spokesperson for the devil. The information processing system described in item 10. [Item 12] The display control unit causes the display device to display a display screen including a plurality of partitioned areas, and controls the display device to display the generated text generated by one of the plurality of artificial intelligences, corresponding to the artificial intelligence, in each of the plurality of areas on the display screen. An information processing system described in any one of items 1 through 11. [Item 13] The display control unit is configured to receive user operations on the multiple areas through the input device and to change the order in which the multiple areas are arranged on the display screen according to the user operations. The information processing system described in item 12. [Item 14] The display control unit causes the display device to display a display screen including a plurality of partitioned areas, and in each of the plurality of areas on the display screen, the plurality of artificial intelligence functions The display device is controlled to display the generated text generated by a corresponding artificial intelligence. Each of the aforementioned multiple areas further displays an input field for writing the input text for the corresponding artificial intelligence. The transmission unit, under the condition that the generated text displayed in the first of the multiple areas is dragged and dropped into the input field of the second of the multiple areas, transmits the dragged generated text to the artificial intelligence corresponding to the input field where the generated text was dropped. The generated sentence acquisition unit is configured to acquire a response sentence for the dragged generated sentence from the artificial intelligence corresponding to the input field where the generated sentence was dropped. The information processing system described in item 1. [Item 15] In each of the aforementioned multiple areas, the content of the dialogue between the user and the corresponding artificial intelligence is displayed by arranging the input sentence by the user and the generated sentence by the corresponding artificial intelligence. The information processing system according to item 14, wherein the transmission unit is configured to transmit the dragged generated sentence, along with a portion of the dialogue content describing the context in which the dragged generated sentence was generated, to the artificial intelligence corresponding to the input field where the generated sentence was dropped. [Item 16] The display control unit controls the display device to display a plurality of display windows, and to display the generated text generated by one of the plurality of artificial intelligences in each of the plurality of display windows, The transmitting unit is configured to determine that the fourth instruction has been input when an operation is performed through the input device to overlay the display window associated with the first artificial intelligence onto the display window associated with the second artificial intelligence, and to cause the first artificial intelligence and the second artificial intelligence to interact. The information processing system described in item 6. [Item 17] The transmitting unit is configured to receive additional instructions from the user via the input device for at least one of the second artificial intelligence and the third artificial intelligence, and is configured to input the additional instructions received by the second artificial intelligence to the second artificial intelligence when transmitting the generated sentence obtained from the first artificial intelligence to the second artificial intelligence, and to input the additional instructions received by the third artificial intelligence to the third artificial intelligence when transmitting the generated sentence obtained from the second artificial intelligence to the third artificial intelligence. The information processing system described in item 8. [Item 18] When the fourth instruction is input, the transmission unit is configured to cause the first artificial intelligence and the second artificial intelligence to converse until at least one of the generated sentences obtained from the first artificial intelligence and the generated sentences obtained from the second artificial intelligence satisfy a predetermined condition, and to terminate the conversation between the first artificial intelligence and the second artificial intelligence once the predetermined condition is satisfied. The information processing system described in item 6. [Item 19] One of the first artificial intelligence and the second artificial intelligence is configured to score the generated sentence input from the other of the first artificial intelligence and the second artificial intelligence. The transmission unit is configured to determine that the predetermined conditions have been met when the score assigned to the generated text by the scoring is equal to or greater than a threshold, and to terminate the dialogue between the first artificial intelligence and the second artificial intelligence. The information processing system described in item 18. [Item 20] A computer program for causing a computer to function as the generated text acquisition unit, the display control unit, and the transmission unit, which are included in the information processing system described in any one of items 1 to 19. [Item 21] A method of information processing performed by a computer, Obtaining generated text from each of multiple artificial intelligences, The generated sentences obtained from each of the aforementioned multiple artificial intelligences are displayed on a display device. The system acquires user input text through an input device and transmits the input text to at least one of the multiple artificial intelligences. Includes, The generated sentence includes a response sentence to the input sentence and is generated by the corresponding artificial intelligence. The aforementioned transmission is, When a first instruction is input from the user through the input device, the input sentence is transmitted to one of the multiple artificial intelligences designated by the user. When a second instruction is input from the user through the input device, the input sentence is transmitted to the plurality of artificial intelligences. Information processing methods including [Explanation of Symbols]

[0230] 10... Information processing system, 11... Processor, 13... Memory, 15... Storage, 19... Communication interface, 60... Information terminal, 61... User interface, 61A... Display device, 61B... Input device, 70... AI server, 71, 71A, 71A_1, 71A_2, 71B... Artificial intelligence, 71M... Master, G0... Service provision screen, G1... Chat screen, G3... Master screen, G11, G31... Profile image, G13, G33... Chat display screen, G15, G16, G35... Input field, G18... Send button, G19... Bulk send button, G37... Operation button group, G371... Send button, G373... Call button, G375... Notepad button, G379... Output button.

Claims

1. A generated text acquisition unit configured to acquire generated text from each of multiple artificial intelligences, A display control unit configured to display the generated sentences obtained from each of the aforementioned multiple artificial intelligences on a display device, A transmission unit configured to acquire input text from a user through an input device and transmit the input text to at least one of the plurality of artificial intelligences, Equipped with, The generated sentence includes a response sentence to the input sentence and is generated by the corresponding artificial intelligence. The aforementioned transmitting unit When a first instruction is input from the user through the input device, the input sentence is transmitted to one of the multiple artificial intelligences designated by the user. If a second instruction is input from the user through the input device, the input sentence is transmitted to the plurality of artificial intelligences. An information processing system configured in such a way.

2. The transmission unit is configured to transmit, when a third instruction is input from the user through the input device, the content of the dialogue between the user and the first artificial intelligence among the plurality of artificial intelligences, through the input sentence and the generated sentence, to a second artificial intelligence among the plurality of artificial intelligences, which is different from the first artificial intelligence. The generated text acquisition unit is configured to acquire, as the generated text, a response text to the dialogue content from the second artificial intelligence. The information processing system according to claim 1.

3. The first artificial intelligence and the second artificial intelligence are each the artificial intelligence specified by the user from among the plurality of artificial intelligences. The information processing system according to claim 2.

4. The transmitting unit is configured to selectively transmit a portion of the dialogue between the user and the first artificial intelligence to the second artificial intelligence when the third instruction is input. The information processing system according to claim 2.

5. The transmission unit is configured to cause the first and second artificial intelligences to interact when a fourth instruction is input from the user through the input device. This is done by transmitting the generated sentence obtained from the first artificial intelligence among the plurality of artificial intelligences to a second artificial intelligence, which is different from the first artificial intelligence, and by transmitting the generated sentence obtained from the second artificial intelligence to the first artificial intelligence. The information processing system according to claim 1.

6. The first artificial intelligence and the second artificial intelligence are each the artificial intelligence specified by the user from among the plurality of artificial intelligences. The information processing system according to claim 5.

7. The transmission unit is configured such that, when a fifth instruction is input from the user through the input device, it transmits the generated sentence obtained from the first artificial intelligence among the plurality of artificial intelligences to a second artificial intelligence, which is different from the first artificial intelligence, and transmits the generated sentence obtained from the second artificial intelligence to a third artificial intelligence, which is different from both the first and second artificial intelligences. The information processing system according to claim 1.

8. The aforementioned transmitting unit The content of the conversation between the user and each of the aforementioned multiple artificial intelligences is transmitted to a specific artificial intelligence that is different from the aforementioned multiple artificial intelligences. If a sixth instruction is input from the user through the input device, the input sentence is transmitted to the specific artificial intelligence. It is configured in such a way, The generated sentence acquisition unit acquires the generated sentence based on the input sentence transmitted to the specific artificial intelligence from the specific artificial intelligence, The display control unit is configured to display the generated sentence obtained from the specific artificial intelligence on the display device. The information processing system according to claim 1.

9. The aforementioned multiple artificial intelligences include two or more artificial intelligences that play different roles. The information processing system according to claim 1.

10. The aforementioned two or more artificial intelligences include an artificial intelligence that acts as a spokesperson for the devil. The information processing system according to claim 9.

11. The display control unit causes the display device to display a display screen including a plurality of partitioned areas, and controls the display device to display the generated text generated by one of the plurality of artificial intelligences, corresponding to the artificial intelligence, in each of the plurality of areas on the display screen. The information processing system according to claim 1.

12. The display control unit is configured to receive user operations on the multiple areas through the input device and to change the order in which the multiple areas are arranged on the display screen according to the user operations. The information processing system according to claim 11.

13. The display control unit causes the display device to display a display screen including a plurality of partitioned areas, and controls the display device to display the generated text generated by one of the plurality of artificial intelligences, corresponding to the artificial intelligence, in each of the plurality of areas on the display screen. Each of the aforementioned multiple areas further displays an input field for writing the input text for the corresponding artificial intelligence. The transmission unit, under the condition that the generated text displayed in the first of the multiple areas is dragged and dropped into the input field of the second of the multiple areas, transmits the dragged generated text to the artificial intelligence corresponding to the input field where the generated text was dropped. The generated sentence acquisition unit is configured to acquire a response sentence for the dragged generated sentence from the artificial intelligence corresponding to the input field where the generated sentence was dropped. The information processing system according to claim 1.

14. In each of the aforementioned multiple areas, the content of the dialogue between the user and the corresponding artificial intelligence is displayed by arranging the input sentence by the user and the generated sentence by the corresponding artificial intelligence. The information processing system according to claim 13, wherein the transmission unit is configured to transmit the dragged generated sentence, along with a portion of the dialogue content describing the context in which the dragged generated sentence was generated, to the artificial intelligence corresponding to the input field where the generated sentence was dropped.

15. The display control unit controls the display device to display a plurality of display windows, and to display the generated text generated by one of the plurality of artificial intelligences in each of the plurality of display windows, The transmitting unit is configured to determine that the fourth instruction has been input when an operation is performed through the input device to overlay the display window associated with the first artificial intelligence onto the display window associated with the second artificial intelligence, and to cause the first artificial intelligence and the second artificial intelligence to interact. The information processing system according to claim 5.

16. The transmitting unit is configured to receive additional instructions from the user via the input device for at least one of the second artificial intelligence and the third artificial intelligence, and is configured to input the additional instructions received by the second artificial intelligence to the second artificial intelligence when transmitting the generated sentence obtained from the first artificial intelligence to the second artificial intelligence, and to input the additional instructions received by the third artificial intelligence to the third artificial intelligence when transmitting the generated sentence obtained from the second artificial intelligence to the third artificial intelligence. The information processing system according to claim 7.

17. When the fourth instruction is input, the transmission unit is configured to cause the first artificial intelligence and the second artificial intelligence to converse until at least one of the generated sentences obtained from the first artificial intelligence and the generated sentences obtained from the second artificial intelligence satisfy a predetermined condition, and to terminate the conversation between the first artificial intelligence and the second artificial intelligence once the predetermined condition is satisfied. The information processing system according to claim 5.

18. One of the first artificial intelligence and the second artificial intelligence is configured to score the generated sentence input from the other of the first artificial intelligence and the second artificial intelligence. The transmission unit is configured to determine that the predetermined conditions have been met when the score assigned to the generated text by the scoring is equal to or greater than a threshold, and to terminate the dialogue between the first artificial intelligence and the second artificial intelligence. The information processing system according to claim 17.

19. A computer program for causing a computer to function as the generated statement acquisition unit, the display control unit, and the transmission unit, which are included in the information processing system according to any one of claims 1 to 18.

20. A method of information processing performed by a computer, Obtaining generated text from each of multiple artificial intelligences, The generated sentences obtained from each of the aforementioned multiple artificial intelligences are displayed on a display device. The system acquires user input text through an input device and transmits the input text to at least one of the multiple artificial intelligences. Includes, The generated sentence includes a response sentence to the input sentence and is generated by the corresponding artificial intelligence. The aforementioned transmission is, When a first instruction is input from the user through the input device, the input sentence is transmitted to one of the multiple artificial intelligences designated by the user. If a second instruction is input from the user through the input device, the input statement , to transmit to the aforementioned multiple artificial intelligences, Information processing methods including

Images