Computer system and execution control method of content

Abstract

A content provision system (1000) has an arrangement information generative artificial intelligence (AI) (20) trained so as to generate arrangement information in accordance with given play data (700) of the content. The content provision system (1000) acquires generation arrangement information (770) by inputting the play data (700) of a user to the arrangement information generative AI (20). The content provision system (1000) changes and arranges application content data (730) on the basis of the acquired generation arrangement information (770), and controls execution of the content.

Description

Computer system and content execution control method

[0001] The present invention relates to a computer system and the like.

[0002] Artificial intelligence (AI) models have made remarkable advances, and various techniques for utilizing AI models in video games have been devised. For example, Patent Literature 1 describes a technique that uses a deep learning model trained with training data that inputs game state information and outputs behavioral information indicating actions to be taken in the game. Furthermore, Patent Literature 2 describes a technique for training an AI model from multiple game plays of a game application scenario using training state data collected from multiple game plays of the scenario and success criteria associated with each of the multiple game plays.

[0003] International Publication No. 2021 / 049254 Japanese Patent Application Laid-Open No. 2022-033164

[0004] Traditionally, video games have been run within the scope of data prepared in advance by the game manufacturer, so the content of the game has remained essentially unchanged. Although some variable elements have been prepared to diversify the content of the game, at first glance it may seem as if the content of the game has changed, in fact the variable content has been predetermined by the game manufacturer.

[0005] For example, when a player selects a game difficulty level before starting play, the strength of enemy NPCs (Non-Player Characters) is set to match the selected game difficulty level. While this may appear to be changing at first glance, the selectable game difficulty levels are actually preset. Therefore, the difficulty level only changes within a pre-designed range. Also, while some games allow players to branch off scenarios and change routes leading to a boss character based on their actions and game performance, both the scenario and the route are limited to a pre-designed range. Even so-called "challenge elements" are limited to the range pre-designed by the game maker.

[0006] The fact that the game content remains essentially unchanged is also advantageous when it comes to gameplay. Because the game content remains essentially unchanged, players can devise ways to beat the game next time based on their previous experience. However, for players who have completed a game, even if they play at a different difficulty level or with the goal of "completing" the game by collecting items or completing all quests, they still wind up repeating the same game content.

[0007] These problems are not limited to video games, but also apply to virtual experience content that provides players with virtual experiences in virtual spaces.

[0008] The problem to be solved by the present invention is to provide a new technology that realizes content that is varied and never gets boring.

[0009] The first disclosure is a computer system that arranges and controls the execution of content played by a user based on given arrangement information, and includes: a generated arrangement information acquisition control unit that acquires generated arrangement information by inputting the user's play data into an arrangement information generation AI (Artificial Intelligence) that has been trained to generate the arrangement information according to given play data of the content; and a content provision control unit that controls the provision of content to the user by controlling the execution of the content arranged based on the generated arrangement information.

[0010] According to the first disclosure, the computer system can acquire generated arrangement information from the arrangement information generation AI and change the content of the content, making it possible to create content that never gets boring by changing the content in various ways.

[0011] The second disclosure is a computer system in which, in the above-mentioned computer system, the content is variably set up and executed based on given content setup information, the arrangement information is modification information for standard setup information related to standard execution control of the content, and the content provision control unit determines the content setup information by modifying the standard setup information based on the generated arrangement information, and sets up and controls execution of the content based on the determined content setup information.

[0012] According to the second disclosure, the computer system can change the content by changing the standard setup information based on the arrange information.

[0013] A third disclosure is the above-mentioned computer system, wherein the content setup information includes setup element information for each setup element, the arrange information includes arrange element information for each setup element to be changed relative to the standard setup information, the arrange information generation AI includes: an element-specific generation AI that generates, for each setup element, the arrange element information for each setup element based on given arrange content; and a front-stage AI that determines the setup element to be changed and the arrange content based on the play data, and controls a generation instruction to cause the element-specific generation AI corresponding to the setup element to generate the arrange element information based on the arrange content, and the content provision control unit determines the setup element information for each setup element based on the arrange element information for each setup element to be changed included in the generated arrange information, and sets up the content.

[0014] According to the third disclosure, the computer system makes it possible to change the content in detail for each setup element.

[0015] The fourth disclosure is a computer system in which, in the above-mentioned computer system, the element-specific generation AI is AI that has been trained to generate the arrangement element information according to the play data and the arrangement content.

[0016] According to the fourth disclosure, the computer system can utilize, as element-specific generation AI, AI that has been trained to generate arrangement element information according to play data and arrangement content.

[0017] A fifth disclosure is a computer system in which, in the above-mentioned computer system, the content is game content, and the setup elements include any of a game stage, a game scenario, a quest, item appearance, parameters of appearing items, an item shop, appearing characters, and parameters of appearing characters.

[0018] According to the fifth disclosure, the computer system is capable of arranging any of the following: game stages, game scenarios, quests, item appearances, parameters of appearing items, item shops, appearing characters, and parameters of appearing characters.

[0019] A sixth disclosure is the computer system, wherein the setup elements include item appearance, and the setup element information includes any of the type of item that appears, the probability of appearance, and a method of acquisition.

[0020] According to the sixth disclosure, the computer system can arrange any of the types of items that appear, the probability of their appearance, and the method of obtaining them.

[0021] A seventh disclosure is the above-mentioned computer system, wherein the setup element information includes a probability of each item appearing, and the computer system further includes an appearance probability presentation control unit that controls presentation of the probabilities to the user.

[0022] According to the seventh disclosure, a computer system can present to a user the probability of each item appearing in content with a given probability.

[0023] An eighth disclosure is the above-mentioned computer system, wherein the setup elements include an item shop, and the setup element information includes any of the type, quantity, and price of items to be sold.

[0024] According to the eighth disclosure, the computer system can arrange the type, quantity, and price of items sold in the item shop.

[0025] The ninth disclosure is a computer system in which the arrangement information generation AI is of multiple types, and the computer system further includes an application AI selection unit that selects an application AI from the multiple types of arrangement information generation AI, and the generation arrangement information acquisition control unit acquires the generation arrangement information by inputting the user's play data into the application AI.

[0026] According to the ninth disclosure, the computer system is able to obtain generated arrangement information using an application AI selected from among multiple types of arrangement information generation AI.

[0027] The tenth disclosure is that, in the above-mentioned computer system, there are multiple types of pre-stage AI, and the computer system further includes an application AI selection unit that selects an application AI from the multiple types of pre-stage AI, and the generation arrangement information acquisition control unit acquires the generation arrangement information by inputting the user's play data into the application AI.

[0028] According to the tenth disclosure, the computer system is able to obtain generated arrangement information using an application AI selected from a plurality of types of pre-stage AIs.

[0029] The eleventh disclosure is a computer system in which, in the above-mentioned computer system, the element-specific generation AI outputs the required time required to generate the arrangement element information in accordance with the generation instruction to the pre-stage AI, and the pre-stage AI recursively determines the setup element to be changed and the arrangement content and controls the generation instruction based on the required time.

[0030] According to the eleventh disclosure, the computer system is able to recursively determine the setup elements to be changed and the arrangement contents and control the generation instructions based on the time required to generate the arrangement element information.

[0031] The twelfth disclosure is a computer system further comprising a past generation information storage control unit that controls the correspondence between past data input to the arrangement information generation AI and past generation arrangement information obtained by the input and stores the data as past generation information, and the generation arrangement information acquisition control unit has an alternative acquisition control unit that controls the reading and acquisition of the past generation arrangement information from the past generation information instead of having the arrangement information generation AI generate the generation arrangement information.

[0032] According to the twelfth disclosure, the computer system can reuse generated arrange information generated in the past, thereby reducing the processing costs and time costs required to generate new arrange information.

[0033] The thirteenth disclosure is a computer system in which, in the above-mentioned computer system, the alternative acquisition control unit controls the reading and acquisition of the past-generated arrangement information corresponding to the past data from the past-generated information when past data exists that satisfies a data approximation condition based on the play data of the user.

[0034] According to the thirteenth disclosure, the computer system can reuse previously generated arrange information that satisfies the data approximation condition.

[0035] The fourteenth disclosure is a computer system in which, in the above-mentioned computer system, the arrangement information generation AI is AI that is trained to analyze annotation target information from the play data and generate the arrangement information based on the analysis results.

[0036] According to the fourteenth disclosure, the computer system is able to obtain generated arrangement information using AI that has been trained using annotation target information from play data.

[0037] The fifteenth disclosure is a computer system in which, in the above-mentioned computer system, the generation arrangement information acquisition control unit acquires the generation arrangement information by inputting the play data of the user playing the content up to a certain point into the arrangement information generation AI, and the content provision control unit arranges and controls the execution of the content from the certain point onwards based on the generation arrangement information.

[0038] According to the fifteenth disclosure, the computer system can acquire generation arrangement information based on the progress of the content, and arrange the content from the middle onwards.

[0039] A sixteenth disclosure is the computer system, wherein the content is multi-play content for multiple users.

[0040] According to the sixteenth disclosure, a computer system can arrange the content of multiplayer content.

[0041] The seventeenth disclosure is a method for a computer system to arrange and control execution of content played by a user based on given arrangement information, the method including: obtaining generated arrangement information by inputting the user's play data into an arrangement information generation AI (Artificial Intelligence) trained to generate the arrangement information according to given play data of the content; and controlling the execution of the content arranged based on the generated arrangement information, thereby controlling content provision to the user.

[0042] 1. A system configuration diagram showing an example configuration of a content providing system. A diagram showing an example data configuration of applicable content data. A diagram for explaining the arrangement of applicable content data. A diagram showing an example data configuration of generated arrangement information. A diagram for explaining arrangement information generation AI. A diagram showing an example of element-specific generation AI. A diagram for explaining past generation information. A diagram for explaining selection of an arrangement information generation AI to be applied. A diagram showing an example game screen. A diagram showing examples of programs and data stored in the server system. A diagram showing an example of functional units implemented on a control board of the server system. A diagram showing an example data configuration of play data. A diagram showing an example data configuration of arrangement control data. A flowchart for explaining the flow of processing related to the provision of one piece of content in the server system. A flowchart continuing from FIG. 14. A flowchart continuing from FIG. 15. A diagram for explaining a modified example. A diagram for explaining a modified example. A diagram for explaining a modified example. A diagram for explaining a modified example.

[0043] Hereinafter, examples of embodiments of the present invention will be described, but it goes without saying that the forms to which the present invention can be applied are not limited to the following embodiments.

[0044] 1 is a system configuration diagram showing an example of the configuration of a content providing system according to this embodiment. The content providing system 1000 is a computer system that controls the execution of a video game (a type of entertainment content that takes place in a virtual world). The content providing system 1000 is a computer system that includes a server system 1100 and a user terminal 1500 for each user, which are connected via a network 9 so as to be able to perform data communications.

[0045] The network 9 refers to a communication path that allows data communication, and includes a dedicated line (dedicated cable) for direct connection, a LAN (Local Area Network) such as Ethernet (registered trademark), a telephone communication network, a cable network, the Internet, and the like.

[0046] The server system 1100 is a computer system that performs various processes such as managing and controlling information of registered users, controlling the progress of the game, etc. The server system 1100 has a main unit 1101 equipped with a control board 1150 .

[0047] The control board 1150 is equipped with various microprocessors such as a CPU (Central Processing Unit) 1151, a GPU (Graphics Processing Unit), and a DSP (Digital Signal Processor), various IC memories 1152 such as a VRAM, RAM, and ROM, and a communication device 1153. Note that some or all of the functions equipped on the control board 1150 may be realized by an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or a SoC (System on a Chip).

[0048] Although the server system 1100 is depicted as if it were a single server device, it may be configured to be realized by a plurality of devices. For example, the server system 1100 may be configured by providing a plurality of servers that share functions and connecting them to each other via an internal bus or the network 9 so that they can communicate data. The server system 1100 may also include a database or online storage.

[0049] The user terminal 1500 serves as a man-machine interface (MMIF) for user 2 to play the game as a player. In other words, for user 2 who is a player, the user terminal 1500 functions as a game play terminal for playing the game. Although only one user terminal 1500 is illustrated, in actual operation, there may be a situation in which multiple user terminals 1500 are connected to the server system 1100 for communication at the same time.

[0050] The user terminal 1500 is a computer system that can be connected to the network 9, such as a personal computer, a smartphone, a wearable computer, a portable game device, a home game device, or a tablet computer.

[0051] The user terminal 1500 is a computer equipped with an operation input device, an image display device, and a control board 1550 that performs arithmetic processing. Examples of the operation input device include a touch panel 1506, a keyboard, a game controller, a mouse, etc. Examples of the image display device include the touch panel 1506, a head-mounted display, a glasses-type display, etc.

[0052] The control board 1550 is equipped with a CPU 1551, various microprocessors such as a GPU and a DSP, various IC memories 1552 such as a VRAM, a RAM and a ROM, a communication module 1553 connected to the network 9, and the like. These elements equipped on the control board 1550 are electrically connected via a bus circuit or the like, and are connected so as to be able to read and write data and send and receive signals. Part or all of the control board 1550 may be an ASIC, an FPGA, or a SoC.

[0053] The control board 1550 stores in the IC memory 1552 programs and various data for realizing the functions of the user terminal 1500. The user terminal 1500 realizes the functions of a play terminal by executing a predetermined application program.

[0054] 2 is a diagram showing an example of the data configuration of the applicable content data 730. The applicable content data 730 is various data that defines the content of the game, and is data for setting up various parameters for controlling the progress of the game.

[0055] The applicable content data 730 is copied from the standard content data 510 before the start of game play and is used in game play. The standard content data 510 is a group of initial setting data prepared in advance by the game maker.

[0056] The applicable content data 730 includes multiple types of content setup information 732 (732a, 732b, ...). One piece of content setup information 732 includes a setup element type and setup element information.

[0057] The "type of setup element" refers to the type of setup element, such as a game stage, a game scenario, a quest, an item appearance, parameters of an item that appears, an item shop, characters that appear, parameters of characters that appear, etc. Of course, other types of setup elements are also acceptable.

[0058] "Setup element information" is a specific information group of setup elements. For example, the setup element types of the content setup information 732b are "game stage" and "stage scenario." The setup element information of the content setup information 732b may be stage order, game space data, background object data, background image data, appearing NPC data, SE (Sound Effect) data, BGM (Background Music) data, etc.

[0059] For example, the setup element type of the content setup information 732d is "appearing character; playable." The setup element information of the content setup information 732d may be character type, character model data that defines the shape, texture data that determines the color scheme of the appearance, motion data that defines the movement, etc. The setup element information may also include ability setting parameter values ​​that define the character's abilities, owned equipment data, spoken voice data, etc.

[0060] 3 is a diagram for explaining the arrangement of the applicable content data 730. For convenience, content setup information 732ba for the first game stage and content setup information 732bb for the second game stage are selectively shown.

[0061] The applied content data 730 is arranged and changed during game play and used to provide content. The initial setup element types and setup element information of the content setup information 732ba and the content setup information 732bb are information copied from the standard content data 510, and this is referred to as "standard setup information." In the standard setup information, the enemy NPC 6a in the first game stage is set to C rank, and the enemy NPC 6b in the second game stage is set to B rank. The B rank enemy NPC 6b is stronger than the C rank enemy NPC 6a.

[0062] The server system 1100 controls the execution of the content provision 10 (10a; play of the first game stage) of the first game stage while referring to the content setup information 732ba. In the example of Figure 3, in the content provision 10 of the first game stage, the player character 4 is fighting an enemy NPC 6 (6a). The server system 1100 stores and manages various information related to the control of the content provision 10 as play data 700.

[0063] The server system 1100 is equipped with an arrangement information generation AI 20, and inputs all or a predetermined type of data from the play data 700 to the arrangement information generation AI 20. The arrangement information generation AI 20 is an AI (Artificial Intelligence) that has been trained to generate arrangement information based on given information annotated from past play data, and outputs generated arrangement information 770 for the content of the game to be played in the future.

[0064] 4 is a diagram showing an example of the data configuration of the generation arrangement information 770. The generation arrangement information 770 is information for specifying the setup details to be realized by arranging, among the content setup information 732 of the applied content data 730. Specifically, the generation arrangement information 770 includes one or more setup change details data 772.

[0065] One set of setup change content data 772 includes a change target setup element type 774 , a change target setup element information type 776 , and change setup information 778 .

[0066] The change setup information 778 is main data specifying the setup content to be realized by the change due to the arrangement. Specifically, the change setup information 778 is information that describes an instruction to replace the setup element information indicated by the change target setup element information type 776 in the content setup information 732 (see FIG. 2 ) indicated by the change target setup element type 774 with the arrangement element information 779.

[0067] 3, during play of the first game stage, the arrangement information generation AI 20 generates arrangement information 770 for the second game stage and onward. Here, it is assumed that arrangement information 770 is generated for the second game stage.

[0068] The server system 1100 executes an arrangement process based on the generated arrangement information 770 for the second game stage, and changes the standard setup information in the content setup information 732bb for the second game stage to modified setup information. In other words, the content setup information 732bb′ is arranged data.

[0069] Enemy NP6b in the second game stage was B rank in the standard setup information, but due to the changes and arrangements made in the modified setup information, it is changed to enemy NPC6c, which is B+ rank (slightly tougher than B rank).

[0070] When play of the first game stage ends, the server system 1100 starts play of the second game stage, providing the content by referencing the arranged content setup information 732bb' (10b). As a result, a B+ rank enemy NPC 6c appears in the second game stage. With this arrangement, if User 2 felt that the game difficulty level due to the initial setup was not enough when playing the first game stage, User 2 will feel a sense of challenge from the second game stage onwards.

[0071] 5 is a diagram for explaining the arrangement information generation AI 20. The arrangement information generation AI 20 includes a pre-stage AI 22 and a plurality of element-specific generation AIs 24.

[0072] The pre-stage AI 22 determines the setup elements to be changed and the arrangement contents indicating the changes to be made to the setup elements to be changed. The "arrangement contents" may be rephrased as the purpose of the arrangement. For example, the arrangement contents in FIG. 3 are "strengthen enemy NPCs."

[0073] Once the setup elements to be changed and the arrangement contents have been determined, the pre-stage AI 22 selects an element-specific generation AI 24 (applicable element-specific generation AI) that generates data for realizing the arrangement contents from among multiple types of element-specific generation AIs 24 that have been prepared in advance.The pre-stage AI 22 then generates and inputs prompts to each element-specific generation AI 24.

[0074] The "prompt" can be set appropriately according to the element-by-element generation AI 24. The prompt includes, for example, a task, a constraint, and reference data to be referred to during generation.

[0075] "Task" indicates the content of the arrangement. In other words, it indicates the purpose of generating the type of data and the background for the generation. "Constraints" indicate restrictions on the data to be generated. For example, constraints include data size, data type, and the types of colors that may be used if image data is to be generated. "Reference data" is data contained in the play data 700 and applicable content data 730 input to the arrangement information generation AI 20. The reference data is the same type of data as the information annotated during the learning process of the element-specific generation AI 24, and is the same type of data as that input during the learning process. The configuration of the prompt and the reference data may vary depending on which element-specific generation AI 24 it is input to.

[0076] 6 is a diagram showing an example of the element-specific generation AI 24. The element-specific generation AI 24 (24a, 24b, ...) is an AI model that has been trained using training data, which takes a prompt as input and outputs setup element information that satisfies the arrangement content indicated in the task of the prompt.

[0077] For example, the first element-specific generation AI 24a is assigned the arrangement element as "game scenario" and the type of arrangement element information as "game world map," and has been trained as a generation AI model that generates a game world map.

[0078] Assume that the prompt to the first element-specific generation AI 24a specifies that the arrangement content indicated in the task is "enlarge the map area by 20%" and that "original map data" is set as the reference data. In this case, the first element-specific generation AI 24a generates and outputs map data that has been expanded by 20% in area relative to the original map data. If no reference data is set, the first element-specific generation AI 24a generates new map data without a draft.

[0079] For example, the second element-specific generation AI 24b has an arrangement element of “appearing character” and the type of arrangement element information is “character model data, character texture data, motion data.” The second element-specific generation AI 24b has already trained as a generation AI model that generates a set of character shape, appearance color scheme, and motion.

[0080] For example, the prompt to the second element-specific generation AI 24b assumes that the arrangement content indicated in the task is "generate a character with a stronger-looking head design" and that "original character data" is set as reference data. In this case, the second element-specific generation AI 24b generates data for a character with, for example, horns on the head that were not present in the original character, a color scheme for the entire character that highlights the horns, and motions that show off or use the horns. If no reference data is set, the second element-specific generation AI 24b generates a new character without a draft.

[0081] In addition, the second element-specific generation AI 24b may be composed of multiple generation AIs based on the type of arrangement element information, such as an AI that generates character model data, an AI that generates texture data, and an AI that generates motion.

[0082] For example, the third element-specific generation AI 24c has an arrangement element of “item appearance” and an arrangement element information type of “appearing item type, appearance probability, ability setting.” The third element-specific generation AI 24c has already trained as a generation AI model that generates item model data, texture data, appearance probability, and item ability setting.

[0083] For example, the prompt to the third element-specific generation AI 24c assumes that the arrangement content indicated in the task is "add a ring item with a common appearance but a powerful effect" and "data of the original ring item" is set as reference data. In this case, the third element-specific generation AI 24c generates data for an item that has a simple design changed from the original ring item, has an unremarkable color scheme, but has ability settings that are far more powerful than the original ring item. If no reference data is set, the third element-specific generation AI 24c will generate a new ring item without a draft.

[0084] The types and number of element-specific generation AIs 24 that the arrangement information generation AI 20 has will vary depending on the learning allocation, that is, how many pieces of setup element information are to be generated.

[0085] To put it in perspective, the pre-stage AI 22 is equivalent to an instructor who decides what arrangement content to apply to each arrangement element, and the element-specific generation AI 24 is equivalent to a craftsman who generates the arrangement elements for which he is responsible.

[0086] Returning to FIG. 5, when the element-by-element generation AI 24 receives a prompt from the preceding AI 22, it predicts the time required to complete generation and returns required time information to the preceding AI 22.

[0087] The pre-stage AI 22 determines whether the prompt needs to be modified based on the returned required time information, and modifies or regenerates the prompt.

[0088] The "determination of whether or not correction is necessary" is divided into two cases: a first case of time leeway until the start of use of the setup element to be changed, and a second case of time leeway until the start of use of the arrangement element information that requires the longest time among the setup element information of the setup element to be changed.

[0089] For example, let us assume that the game stage currently being played is the first game stage, the setup element to be changed is the second game stage, and the setup element information requiring the longest time is "game space data." This data is the data to be used from the beginning of play of the second game stage. The time difference between the elapsed time from the start of play and the expected time to clear the first game stage is the first time margin.

[0090] If the longest required time exceeds the first time margin under this assumption, the system cannot wait for the completion of the generation of the setup element information requiring the longest required time. Therefore, under this assumption, it is determined that correction is "necessary," and the front-stage AI 22 corrects the prompt related to the generation of the setup element information requiring the longest required time so as to shorten the required time. Alternatively, the front-stage AI 22 regenerates the prompt, acquires the required time information again, and repeats the determination of whether correction is required until it determines that correction is "not required."

[0091] On another premise, assume that the game currently being played is the first game stage, and the setup element to be changed is the second game stage, but the setup element information that requires the longest time is the "enemy boss NPC data" that appears at the end of the stage. This data is used at the end of the second game stage.

[0092] Under this premise, if the sum of the first time allowance and the second time allowance from the start of the second game stage until the enemy boss NPC appears exceeds the longest required time, it is possible to play the second game stage while waiting for the generation of the enemy boss NPC data to be completed.

[0093] Therefore, if the sum of the first time margin and the second time margin is less than the longest required time, it is determined that modification is "necessary," and the pre-stage AI 22 modifies the prompt related to the generation of the setup element information requiring the longest required time so as to shorten the required time. Alternatively, the arrangement information generation AI 20 modifies the prompt for generating the stage scenario (scenario script) that determines the flow of the game stage so that the timing of the appearance of the enemy boss NPC is later than that of the standard setup. Of course, the pre-stage AI 22 may regenerate the prompt rather than modify it, obtain the required time information again, and repeat the determination of whether modification is required until it determines that modification is "not required."

[0094] If it is determined that no modification is required, the pre-stage AI 22 inputs a generation command to the element-specific generation AI 24 that received the prompt. The element-specific generation AI 24 generates arrangement element information according to the acquired prompt. At this point, the arrangement information generation AI 20 outputs generated arrangement information 770 that associates the type of arrangement element to be changed with the generated arrangement element information.

[0095] 7 is a diagram for explaining the past generation information 800. The server system 1100 associates a prompt generated by the arrangement information generation AI 20 with the generated arrangement information 770 generated based on the prompt, and stores the association as the past generation information 800. Specifically, one piece of past generation information 800 stores a past generation information ID 802, a past prompt 804 which is a copy of the generated prompt, and past generation arrangement information 806 which is a copy of the generated arrangement information 770.

[0096] A prompt that is defined as a past prompt 804 includes a task, constraints, and reference data. The reference data is data extracted from the play data 700 input to the arrangement information generation AI 20 when the prompt was generated, or from the applicable content data 730. Therefore, the reference data of the past prompt 804 includes past play data at the time the prompt was generated.

[0097] After generating a new prompt, the arrangement information generation AI 20 searches through the past generated information 800 for past prompts 804 that have the same content as the newly generated prompt or are similar in content to the newly generated prompt by satisfying specified data approximation conditions.

[0098] The "data approximation conditions" are conditions that must be met to allow reuse of the generated arrangement information 770, and are conditions for determining whether a newly generated prompt is similar to a past prompt. Therefore, the data approximation conditions are written and defined in advance as, for example, conditions regarding differences in task content or conditions regarding differences in constraints.

[0099] As described above, the tasks and constraints included in the prompts are generated based on the play data 700 that represents the original state to be arranged, and the reference data included in the prompts is excerpted data from the play data 700 and the applicable content data 730. In light of this, satisfying the data approximation conditions means that "there is (was) past play data that satisfies the data approximation conditions based on the play data."

[0100] The past generation information 800 may further include other information, for example, a copy of the setup change data 772 in Fig. 4 as past setup change data. This past setup change data may then be used to determine the data approximation conditions.

[0101] If the arrange information generation AI 20 finds reusable past generated information 800, it reads out that past generated arrange information 806. Then, the arrange information generation AI 20 outputs the read out past generated arrange information 806 as the generated arrange information 770 (7, 70a) instead, without issuing a generation instruction related to the generated prompt.

[0102] If the corresponding past generation information 800 cannot be found, the arrangement information generation AI 20 issues a generation instruction as described above to the element-specific generation AI 24 to execute generation, and outputs new generated arrangement information 770 (770b).

[0103] The time required and the computational resources can be reduced by reusing the past generation arrangement information 770. If the time required can be reduced, the need to modify and regenerate prompts can also be avoided.

[0104] 8 is a diagram for explaining the selection of the arrangement information generation AI 20 to be applied. Multiple kinds and types of arrangement information generation AI 20 are prepared in advance, and which of them will be applied when the content is provided is determined based on the setting operation performed by the player before the start of the game.

[0105] Specifically, before the game starts, the server system 1100 displays a setting screen W2 on the user terminal 1500 to accept setting operations for arrangements. The setting screen W2 has a function ON / OFF setting section 30 and an arrangement type selection section 32.

[0106] The ON / OFF setting unit 30 accepts a selection setting of whether to enable (ON) or disable (OFF) the function itself for arranging the content during game play.

[0107] The arrangement type selection unit 32 displays arrangement type explanation text 34 corresponding to each type of arrangement information generation AI 20 in association with a selection operation switch 36. The arrangement type explanation text 34 succinctly expresses the direction of arrangement of the arrangement information generation AI 20. In this embodiment, the arrangement type selection unit 32 can select one type, but it may also be configured to allow multiple selections.

[0108] In the example of Figure 8, the type "high difficulty level for a serious challenge" is selected in the arrangement type selection section 32, and the arrangement information generation AI 20 (20a) corresponding to that type is set as the applicable arrangement information generation AI 29.

[0109] Each arrangement information generation AI 20 (20a, 20b, ...) is set with a notification character 7 (7a, 7b, ...) of a different design. When the server system 1100 executes an arrangement process during content play, an image of the notification character 7 corresponding to the applied arrangement information generation AI 29 appears on the screen together with an arrangement execution notification 8, as shown in the example game screen W4 of Figure 9. The example of Figure 9 is an example game screen when the arrangement information generation AI 20 (20d) of Figure 8 is selected as the applied arrangement information generation AI 29, and the notification character 7 (7d) and the arrangement execution notification 8 are displayed.

[0110] The notification character 7 acts as a director who directed the arrangement or an announcer who notifies the player of the execution of the arrangement, and notifies the player of the details of the arrangement. The arrangement execution notification 8 displays text that succinctly announces the details of the arrangement. The contents of the arrangement execution notification 8 can be set as appropriate, but if an arrangement related to the appearance of an item has been executed, it should include information about the item appearance rate.

[0111] 10 is a diagram showing examples of programs and data stored in the server system 1100. The server system 1100 stores in the IC memory 1152 a server program 501, a distribution client program 502, standard content data 510, notification character definition data 570, and arrangement information generation AI model data 590. The server system 1100 also stores arrangement execution condition definition data 600, play data 700, arrangement control data 750, past generation information 800, and current date and time 900. Of course, data other than these may also be stored as appropriate.

[0112] The server system 1100 executes and processes the server program 501 in the CPU 1151, thereby realizing the function of a server processing unit 200s in the control board 1150, as shown in FIG. 11, for example.

[0113] The server processing unit 200s has an application AI selection unit 208, a generation arrangement information acquisition control unit 210, an arrangement information generation AI execution unit 218, a past generation information storage control unit 220, a content provision control unit 222, an appearance probability presentation control unit 224, and a timing unit 280.

[0114] The application AI selection unit 208 selects an application AI (application arrangement information generation AI 29) from among a plurality of types of arrangement information generation AIs 20 (20a, 20b, . . . ; see FIG. 8).

[0115] The generated arrangement information acquisition control unit 210 acquires generated arrangement information 770 by inputting the user's play data into the arrangement information generation AI 20, which has been trained to generate arrangement information according to given play data of the content.

[0116] The generated arrange information acquisition control unit 210 also has an alternative acquisition control unit 212. The alternative acquisition control unit 212 controls the arrangement information generation AI 20 to read and acquire past generated arrange information 806 from the past generated information 800, instead of causing the arrangement information generation AI 20 to generate the generated arrange information 770 (see FIG. 8).

[0117] The arrangement information generation AI execution unit 218 is a functional unit corresponding to each of the arrangement information generation AI model data 590. The arrangement information generation AI execution unit 218 executes the corresponding arrangement information generation AI model data 590.

[0118] The past generation information storage control unit 220 controls the past play data input to the arrangement information generation AI 20 to be associated with the past generated arrangement information acquired by the input, and stores the associated data as past generation information 800 (see FIG. 7).

[0119] The content provision control unit 222 controls the provision of content to the user by controlling the execution of the content arranged based on the generated arrangement information 770. Specifically, the content provision control unit 222 determines the content setup information 732 by changing the standard setup information based on the generated arrangement information 770. Then, the content provision control unit 222 sets up the content based on the determined content setup information 732 and controls its execution.

[0120] More specifically, the content provision control unit 222 determines setup element information for each setup element based on the arrangement element information for each setup element to be changed included in the generated arrangement information 770, and sets up the content.

[0121] Incidentally, since the content in this embodiment is a video game, the content provision control unit 222 functions as a game progress control unit 223. The game progress control unit 223 prepares applicable content data 730, sets up play data 700 based on that data, and executes processing related to game progress. The game progress control unit 223 then executes processing such as updating the play data 700 in accordance with the game progress and determining game results.

[0122] The appearance probability display control unit 224 controls displaying the probability of each item appearing with a given probability in the content to the user (player), as in the example of the arrangement execution notification 8 in FIG.

[0123] The timekeeping unit 280 uses a system clock to measure various times such as the current date and time 900 (see FIG. 12), the time limit for a mission, and the game time.

[0124] Returning to FIG. 10, the distribution client program 502 is the original client program provided to the user terminal 1500 .

[0125] The notification character definition data 570 is prepared for each notification character 7 (see FIG. 9).

[0126] Arrangement information generation AI model data 590 is prepared for each type of arrangement information generation AI 20. One arrangement information generation AI model data 590 includes an arrangement type 592 corresponding to the type of AI model, an input data type list 593, pre-stage AI model data 594, element-specific generation AI model data 596, and an element-specific reference data type list 597. Of course, data other than these may also be included as appropriate.

[0127] The input data type list 593 indicates which type of data from the play data 700 is to be input to the arrangement information generation AI 20 for the model data. The data indicated by this list is the same type as the data annotated and input during learning of the arrangement information generation AI 20. Note that this list may differ depending on the arrangement type 592 of the arrangement information generation AI 20.

[0128] The element-specific reference data type list 597 is prepared in one-to-one correspondence with the element-specific generated AI model data 596. The element-specific reference data type list 597 specifies which type of data is to be selected from the data included in the play data 700 and the applicable content data 730 as reference data to be input to the corresponding AI model.

[0129] The arrangement execution condition definition data 600 is data that defines the conditions that must be met in order for the arrangement information generation AI 20 to generate the generated arrangement information 770 and execute the arrangement process, and is prepared for each arrangement type 592 of the arrangement information generation AI 20.

[0130] The arrange execution condition definition data 600 stores an applicable arrange type indicating the arrange information generation AI 20 to which the definition data applies, condition description data describing the arrange execution conditions, and a data type for determining whether the conditions are met. The data type for determining whether the conditions are met specifies which type of data in the play data 700 is to be used to determine whether the arrange execution conditions are met or not.

[0131] The "arrangement execution condition" may be, for example, the end of a game stage currently being played. The arrangement execution condition may also be set appropriately, such as the end of a quest, when a player's parameter value satisfies a predetermined condition, or every time a predetermined amount of game time has elapsed.

[0132] For example, the arrangement execution conditions for the arrangement information generation AI 20a (see FIG. 8) that adjusts the difficulty level to a higher level (more difficult) may be described in the condition description data as the rate of decrease in the player's HP per unit time or the rate at which the player is hit. In this case, the arrangement execution conditions can detect a situation in which the difficulty level of the content indicated by the currently applied content data 730 is too high for the player's playing skill, and generate the arrangement information 770 and execute the arrangement process.

[0133] In this way, the arrangement execution conditions must be set appropriately depending on the direction of arrangement, that is, the purpose of arrangement, of the arrangement information generation AI 20 to which the conditions are applied.

[0134] Play data 700 is prepared for each play of content and stores various data describing the status of provision of that content. For example, as shown in Figure 12, one play data 700 includes a play ID 702, a player account 704, game status information 710, and applicable content data 730 (see Figure 2). Of course, other data may also be included as appropriate.

[0135] The game status information 710 is provision status information 711 that indicates what kind of content is being provided. For example, the game status information 710 includes a game time 712, a currently played stage ID log 713, a currently played quest ID log 714, a player character status log 715, an NPC status log 716, and a game performance data log 717. The "log" is a record that associates data with a date and time, and is history information.

[0136] 13 is a diagram showing an example of the data structure of arrangement control data 750. Arrangement control data 750 is prepared for each play data 700. One arrangement control data 750 includes a play ID 751 indicating the corresponding play data 700, an applicable arrangement information generation AI type 752, arrangement content setting data 754, an applicable element-specific generation AI type list 755, and prompt setting data 756. The arrangement control data 750 also includes required time information 758 and generation arrangement information 770 (see FIG. 4). Of course, other data may also be included as appropriate.

[0137] The applied arrange information generation AI type 752 indicates the applied arrange information generation AI 29 in FIG.

[0138] The arrangement content setting data 754 is created for each setup element to be changed. One arrangement content setting data 754 includes the type of setup element to be changed and arrangement content description data that describes the arrangement content.

[0139] The prompt setting data 756 is prepared for each element-specific generation AI 24 that inputs a prompt. One prompt setting data 756 stores the type of element-specific generation AI to be input and prompt data.

[0140] The required time information 758 is prepared for each element-based generated AI 24 for which a prompt has been input. One piece of required time information 758 stores the type of element-based generated AI and the predicted required time length.

[0141] 14 to 16 are flowcharts for explaining the flow of processing related to the provision of one piece of content in the server system 1100. As a premise, the arrange function is selected to "ON" in the ON / OFF setting section 30 on the setting screen W2 (see FIG. 8).

[0142] As shown in Figure 14, the server system 1100 first sets the players (step S10). The number of players may be one or more. The server system 1100 then accepts the setting operation for the arrangement function and determines the applicable arrangement information generation AI 29 (step S12; see Figure 8). In the case of a multiplayer game, step S12 applies to one of the players and allows the representative player to input.

[0143] Next, the server system 1100 copies the standard content data 510 (see FIG. 10) and sets the applicable content data 730 (see FIG. 12) (step S14), thereby initially setting up the content.

[0144] Next, the server system 1100 references the applicable content data 730, initializes the game space of the game stage to be played first (step S16), and starts controlling the provision of content (step S18). At the same time, the server system 1100 starts recording and updating the play data 700 as needed.

[0145] When content provision control (game play progress control) begins, the server system 1100 refers to the arrangement execution condition definition data 600 applied to the applied arrangement information generation AI 29, and monitors whether the arrangement execution conditions are met.

[0146] If the arrangement execution conditions are met (YES in step S30), the server system 1100 selects data of the type indicated in the input data type list 593 (see Figure 10) from the play data 700 and inputs it to the applied arrangement information generation AI 29 (step S32).

[0147] The applicable arrange information generation AI 29 determines the setup element to be changed and the arrange content, and creates arrange content setting data 754 (see FIG. 13) (step S36). Next, the applicable arrange information generation AI 29 selects, from among a plurality of element-specific generation AIs 24 (see FIG. 6), the element-specific generation AI 24 corresponding to the determined setup element to be changed as the applicable element-specific generation AI (step S38). The selection result is stored as the applicable element-specific generation AI type list 755 (see FIG. 13).

[0148] Next, the applicable arrangement information generation AI 29 generates a prompt for each applicable element generation AI (step S40). Then, it searches for previously generated information 800 that has a past prompt 804 that satisfies the newly generated prompt and the data approximation condition, i.e., previously generated information 800 that can be reused (step S42; see FIG. 7).

[0149] 15, if there is previously generated information 800 that can be reused (YES in step S60), the applicable arrange information generation AI 29 acquires previously generated arrange information 806 from the corresponding previously generated information 800. Then, the applicable arrange information generation AI 29 outputs the acquired previously generated arrange information 806 as an alternative to generating new generated arrange information 770 (step S62).

[0150] Next, when the generated arrange information 770 is output from the applied arrange information generation AI 29, the server system 1100 executes the arrange process based on this (step S64) and notifies the execution of the arrange within the content (step S66; see FIG. 9).

[0151] On the other hand, if there is no previously generated information 800 that can be reused (NO in step S60), the applicable arrangement information generation AI 29 inputs a prompt to the element-specific generation AI 24 (step S70) and obtains the required time information 758 (step S72).

[0152] The applicable arrange information generation AI 29 determines whether the prompt needs to be modified (step S74). If it determines that the prompt needs to be modified (NO in step S74), the applicable arrange information generation AI 29 regenerates or modifies the prompt (step S76).

[0153] Once the prompt is regenerated or modified, the process returns to step S42, and if there is previously generated information 800 that can be reused for that prompt (YES in step S60), the applicable arrange information generation AI 29 acquires the previously generated arrange information 806 instead (step S62). However, if the regenerated prompt is again determined to "require" modification (NO in step S74), the applicable arrange information generation AI 29 regenerates or modifies the prompt once more and returns to step S42.

[0154] If the prompt modification necessity determination result is "no modification required" during this recursive process (YES in step S74), the applicable arrange information generation AI 29 inputs a generation instruction to the element-specific generation AI 24 (step S80). The element-specific generation AI 24 executes generation based on the input prompt, and the applicable arrange information generation AI 29 generates and outputs new generated arrange information 770 (step S82).

[0155] When the generated arrange information 770 is output from the applied arrange information generation AI 29, the server system 1100 executes the arrange process based on the generated arrange information 770 (step S84), notifies the execution of the arrange process within the content (step S86), and then stores the new past generated information 800 (step S88).

[0156] 16, the server system 1100 determines whether or not the play data 700 satisfies a predetermined end condition (step S102), that is, whether or not the game end condition has been satisfied.

[0157] If the content has not ended (NO in step S100), the server system 1100 determines whether the game stage or quest being played has ended (step S102), and if so, switches the game stage or quest being played (step S104). If this switching causes the provision of content based on the setup elements changed from the standard setup information to the changed setup information, provision of the arranged content will begin.

[0158] If the content has ended (YES in step S100), the server system 1100 executes a predetermined game ending process (step S110) and ends the series of processes.

[0159] The game end processing may include announcing the game results, saving data for restarting the game, generating and saving training data for additional learning by the application arrangement information generation AI 29 from past generation information 800 (see Figure 7), and so on.

[0160] The training data for additional learning is generated by analyzing the annotation target information (data indicated in the input data type list 593 of the applicable arrangement information generation AI 29; see FIG. 10 ) in the play data 700. When training data for additional learning is generated, the server system 1100 may use the training data to perform additional learning on the applicable arrangement information generation AI 29.

[0161] Of course, if the arrange function is selected and set to "OFF" in the ON / OFF setting section 30 on the setting screen W2 (see FIG. 8), steps S30 to S88 are omitted. In this case, the content remains unchanged from the standard setup information, and the game is executed with the content previously set by the game maker, just like a conventional video game.

[0162] As described above, this embodiment makes it possible to realize a new technology for realizing content that can be varied in various ways to provide a virtual experience that never gets boring. That is, the content providing system 1000 can acquire generated arrangement information using the arrangement information generation AI 20 and change the content of the content. This provides players with a game that never gets boring, and allows game developers to avoid having to repeatedly update the game after release.

[0163] [Modifications] Although examples of embodiments to which the present invention is applied have been described above, the forms to which the present invention can be applied are not limited to the above-described forms, and components can be added, omitted, or modified as appropriate.

[0164] (Variation 1) The content providing system 1000 is not limited to a client-server system. As shown in FIGS. 17 and 18, the content providing system 1000B may be a standalone user terminal 1500B. In this case, the user terminal 1500B executes a game program 503 (see FIG. 17) to realize the function of the processing unit 200 (see FIG. 18). The processing unit 200 has the same functional configuration as the server processing unit 200s (see FIG. 11). With this configuration, the same effects as those of the above embodiment can be obtained.

[0165] Furthermore, for example, the content providing system 1000 may be realized by a P2P (Peer to Peer) architecture for multiple user terminals 1500. In this case, programs and data according to the division of functions are stored in the user terminals 1500, and the functions of the processing unit 200 are distributed and realized by the user terminals 1500 that serve as P2P nodes. With this configuration, the same effects as those of the above embodiment can be obtained.

[0166] (Variation 2) The genre of the game that is the content executed by the content providing system 1000 can be set as appropriate. Furthermore, the content is not limited to a game as long as it provides a virtual experience and the content can change as the user plays. For example, the content may be a virtual experience of playing in a virtual theme park, a virtual activity experience such as skydiving or scuba diving, or a virtual live performance experience of a virtual idol.

[0167] (Variant 3) In the above embodiment, an example was given of a configuration in which the arrangement information generation AI 20 includes a pre-stage AI 22 and an element-specific generation AI 24, but it may also be a configuration in which some or all of the element-specific generation AI 24 is separate from the arrangement information generation AI 20 and is called and used as an external service.

[0168] 19 shows an example of a configuration in which all of the element-specific generation AIs 24 are provided by an external server system 1100G. There may be an external server system 1100G for each element-specific generation AI 24, or one external server system 1100G may provide multiple element-specific generation AIs 24.

[0169] In this configuration, the pre-stage AI 22 calls the element-specific generation AI 24 to be used using the respective APIs (Application Programming Interfaces) to input prompts, acquire required time information, and acquire generated data. The arrangement information generation AI 20 then packages the acquired generated data and outputs generated arrangement information 770.

[0170] (Modification 4) In the above embodiment, the arrangement information generating AI 20 has one unique pre-stage AI 22, which gives the arrangement its own individuality. However, the present invention is not limited to this.

[0171] 20 includes a plurality of pre-stage AIs 22 (22a, 22b, ...) each corresponding to one of the arrange types.

[0172] The pre-stage AI 22 corresponding to the arrangement type selected and set in the arrangement type selection section 32 of the setting screen W2 is set as the applied pre-stage AI 28 and applied to the generation of the generated arrangement information 770. In this configuration, the applied AI selection section 208 (see Figure 11) performs the function of setting the applied pre-stage AI 28.

[0173] 10...Content provision 20...Arrangement information generation AI 22...Previous stage AI 24...Element-specific generation AI 28...Applied previous stage AI 29...Applied arrange information generation AI 208...Applied AI selection unit 210...Generated arrange information acquisition control unit 212...Alternative acquisition control unit 220...Past generation information storage control unit 222...Content provision control unit 224...Appearance probability presentation control unit 501...Server program 510...Standard content data 590...Arrangement information generation AI model data 593...Input data type list 594...Previous stage AI model data 596...Element-specific generation AI model data 597...Element-specific reference data type list 600...Arrangement execution condition definition data 700...Play data 730...Applied content data 732...Content setup information 750...Arrangement control data 752...Applied arrange information generation AI type 754...Arrangement content setting data 755... List of generated AI types by applicable element 756... Prompt setting data 758... Required time information 770... Generated arrangement information 772... Setup change content data 774... Type of setup element to be changed 776... Type of setup element information to be changed 778... Changed setup information 779... Arrangement element information 800... Past generation information 1000... Content providing system 1100... Server system 1500... User terminal

Claims

1. A computer system that arranges and controls the execution of content played by a user based on given arrangement information, comprising: a generated arrangement information acquisition control unit that acquires generated arrangement information by inputting the user's play data into an arrangement information generation AI (Artificial Intelligence) that has been trained to generate the arrangement information in accordance with given play data of the content; and a content provision control unit that controls the provision of content to the user by controlling the execution of the content arranged based on the generated arrangement information.

2. The computer system of claim 1, wherein the content is variably set up and executed based on given content setup information, the arrangement information is modification information for standard setup information related to standard execution control of the content, and the content provision control unit determines the content setup information by modifying the standard setup information based on the generated arrangement information, and sets up and controls execution of the content based on the determined content setup information.

3. The computer system according to claim 2, wherein the content setup information includes setup element information for each setup element, the arrange information includes arrange element information for each setup element to be changed relative to the standard setup information, the arrange information generation AI comprises: an element-specific generation AI that generates, for each of the setup elements, the arrange element information for each of the setup elements based on given arrange contents, and a front-stage AI that determines the setup element to be changed and the arrange contents based on the play data, and controls generation instructions to cause the element-specific generation AI corresponding to the setup element to be changed to generate the arrange element information based on the arrange contents, and the content provision control unit determines the setup element information for each of the setup elements based on the arrange element information for each of the setup elements to be changed included in the generated arrange information, and sets up the content.

4. The computer system according to claim 3, wherein the element-specific generation AI is AI trained to generate the arrangement element information in accordance with the play data and the arrangement content.

5. The computer system according to claim 3 or 4, wherein the content is game content, and the setup elements include any of a game stage, a game scenario, a quest, item appearance, parameters of appearing items, an item shop, appearing characters, and parameters of appearing characters.

6. The computer system according to claim 5, wherein the setup elements include item appearance, and the setup element information includes any one of the type of item that appears, the probability of appearance, and how to obtain the item.

7. The computer system according to claim 6, wherein the setup element information includes the probability of each item appearing, and further comprising an appearance probability presentation control unit that controls the presentation of the probabilities to the user.

8. The computer system according to any one of claims 5 to 7, wherein the setup elements include an item shop, and the setup element information includes any one of the type, quantity, and price of items to be sold.

9. A computer system as described in any one of claims 1 to 8, further comprising: an application AI selection unit that selects an application AI from among the plurality of types of arrangement information generation AI; and the generation arrangement information acquisition control unit that acquires the generation arrangement information by inputting the user's play data into the application AI.

10. A computer system as described in any one of claims 3 to 8, further comprising: an application AI selection unit that selects an application AI from among the plurality of types of pre-stage AI; and the generation arrangement information acquisition control unit that acquires the generation arrangement information by inputting the user's play data into the application AI.

11. The computer system described in any one of claims 3 to 8, wherein the element-specific generation AI outputs the required time required to generate the arrangement element information in accordance with the generation instruction to the pre-stage AI, and the pre-stage AI recursively determines the setup element to be changed and the arrangement content and controls the generation instruction based on the required time.

12. A computer system as described in any one of claims 1 to 11, further comprising a past generation information storage control unit that controls the association of past data input to the arrangement information generation AI with past generated arrange information obtained by said input and storing the associated data as past generated information, wherein the generated arrange information acquisition control unit has an alternative acquisition control unit that controls the arrangement information generation AI to read and acquire the past generated arrange information from the past generated information instead of having the arrangement information generation AI generate the generated arrange information.

13. The computer system described in claim 12, wherein the alternative acquisition control unit performs control to read and acquire the previously generated arrangement information corresponding to the past data from the previously generated information when there is past data that satisfies the data approximation conditions based on the play data of the user.

14. A computer system according to any one of claims 1 to 13, wherein the arrangement information generation AI is AI trained to analyze annotation target information from the play data and generate the arrangement information based on the analysis results.

15. A computer system as claimed in any one of claims 1 to 14, wherein the generation arrangement information acquisition control unit acquires the generation arrangement information by inputting the play data of the user playing the content up to a certain point into the arrangement information generation AI, and the content provision control unit arranges and controls the execution of the content from the certain point onwards based on the generation arrangement information.

16. A computer system according to any one of claims 1 to 15, wherein the content is multi-play content for multiple users.

17. A method for a computer system to arrange content played by a user based on given arrangement information and control its execution, the method comprising: obtaining generated arrangement information by inputting the user's play data into an arrangement information generation AI (Artificial Intelligence) trained to generate the arrangement information in accordance with given play data of the content; and controlling the provision of content to the user by controlling the execution of the content arranged based on the generated arrangement information.

Images