Electronic device, method, and non-transitory computer-readable storage medium for generating message

Abstract

An electronic device according to an embodiment may identify an event for generating a response. The electronic device may, on the basis of identifying the event, generate a prompt for replacing information, related to a second storage area different from a first storage area of memory, with a keyword. The electronic device may provide the prompt to a trained model configured to be able to access the first storage area among the first storage area and the second storage area of the memory. The electronic device may obtain a first response, generated according to the prompt, from the trained model. The electronic device may generate a second response by replacing the keyword, which is included in the first response, with privacy information in the second storage area of the memory.

Description

Electronic device, method, and non-transitory computer-readable storage medium for generating a message

[0001] The present disclosure relates to an electronic device, a method, and a non-transitory computer-readable storage medium for generating a message.

[0002] Electronic devices can provide various types of content (e.g., text, images). These contents can be received from other electronic devices, or they can be generated by the electronic device and then provided to the user or transmitted to another electronic device.

[0003] For example, software applications and / or services utilizing artificial intelligence are being distributed. By executing the software applications and / or services, users can obtain content in the form of at least one of natural language (e.g., text and / or audio signals), images, and / or videos.

[0004] The above information is provided solely as background information to aid in understanding the present disclosure. No determination has been made, nor is any assertion made, that any of the above constitutes prior art with respect to the present disclosure.

[0005] Aspects of the present disclosure exist to at least address the problems and / or disadvantages described above, and to at least provide the advantages described below. Accordingly, aspects of the present disclosure exist to provide an electronic device, method, and non-transitory computer-readable storage medium for generating a message.

[0006] Additional aspects will be set forth in part by the description which follows, and in part will be obvious from the description, or may be learned by practice of the disclosed embodiments.

[0007] According to one aspect of the present disclosure, an electronic device is provided. The electronic device may include at least one processor, the electronic device including one or more storage media, a memory storing instructions, and a processing circuit. The instructions, when individually or collectively executed by the at least one processor, may cause the electronic device to identify an event for generating a response. The instructions, when individually or collectively executed by the at least one processor, may cause the electronic device to generate a prompt for replacing information associated with a second storage area different from a first storage area of ​​the memory with a keyword based on identifying the event. The instructions, when individually or collectively executed by the at least one processor, may cause the electronic device to provide the prompt to a trained model within the electronic device, the first storage area being accessible from among the first storage area or the second storage area of ​​the memory. The instructions, when individually or collectively executed by the at least one processor, may cause the electronic device to obtain a first response generated according to the prompt from the trained model. The instructions, when individually or collectively executed by the at least one processor, may cause the electronic device to generate a second response by replacing the keyword included in the first response with privacy information in the second storage area of ​​the memory.

[0008] In one embodiment, an electronic device may include at least one processor, the processor including one or more storage media, a memory storing instructions, and a processing circuit. The instructions, when individually or collectively executed by the at least one processor, may cause the electronic device to provide a conversation session in which at least one message exchanged between a first user of the electronic device and a second user of another electronic device is displayed. The instructions, when individually or collectively executed by the at least one processor, may cause the electronic device to, in response to a first request, input the at least one message into a first model configured to generate a string based on provided input, and obtain a first message from the first model for transmission to the other electronic device. The instructions, when individually or collectively executed by the at least one processor, may cause the electronic device to display a second message in which the designated identifier is replaced with privacy information stored in the electronic device, at least in part based on a determination that the first message includes a designated identifier. The instructions, when individually or collectively executed by the at least one processor, may cause the electronic device to, in response to a second request, transmit the second message to the other electronic device through the conversation session.

[0009] In one embodiment, a method of an electronic device may be provided. The method may include providing a conversation session in which at least one message exchanged between a first user of the electronic device and a second user of another electronic device is displayed. The method may include, in response to a first request, inputting the at least one message into a first model configured to generate a string based on the provided input, thereby obtaining a first message for transmission to the other electronic device from the first model. The method may include, based at least in part on a determination that the first message includes a designated identifier, displaying a second message in which the designated identifier is replaced with privacy information stored in the electronic device. The method may include, in response to a second request, transmitting the second message to the other electronic device through the conversation session.

[0010] In one embodiment, a non-transitory computer-readable storage medium storing instructions may be provided. The instructions, when executed by an electronic device including a memory, may cause the electronic device to identify a user request. The instructions, when executed by the electronic device, may cause the electronic device to generate a response to the user request and to generate a prompt for replacing inaccessible information to be included in the response with a keyword. The instructions, when executed by the electronic device, may cause the electronic device to provide the prompt to a trained model within the electronic device configured to access a first storage area of ​​the memory. The instructions, when executed by the electronic device, may cause the electronic device to obtain a first response generated in response to the prompt from the trained model. The instructions, when executed by the electronic device, may cause the electronic device to generate a second response by replacing the keyword included in the first response with privacy information in a second storage area of ​​the memory.

[0011] In one embodiment, a non-transitory computer-readable storage medium storing instructions may be provided. The instructions, when executed by an electronic device including a memory, may cause the electronic device to perform operations. The operations may include providing a chat session in which at least one message is displayed to be exchanged between a first user of the electronic device and a second user of another electronic device; inputting the at least one message into a first model configured to generate character strings based on provided input, thereby obtaining a first message to be transmitted from the first model to the other electronic device in response to a first request; displaying a second message in which the designated identifier is replaced with privacy information stored in the electronic device, based at least on a determination that the first message includes a designated identifier; and transmitting the second message to the other electronic device through the chat session based on a second request.

[0012] Other aspects, advantages, and salient features of the present disclosure will become apparent to those skilled in the art from the following detailed description of various embodiments of the present disclosure, taken in conjunction with the accompanying drawings.

[0013] The above and other aspects, features, and advantages of some embodiments of the present disclosure will become more apparent from the following description taken in conjunction with the accompanying drawings, in which:

[0014] FIG. 1 illustrates the operation of an electronic device, in response to a user request, to generate a response including privacy information, according to one embodiment of the present disclosure;

[0015] FIG. 2 illustrates a block diagram of an electronic device according to one embodiment of the present disclosure;

[0016] FIG. 3 illustrates a flow diagram of an electronic device according to one embodiment of the present disclosure;

[0017] FIG. 4 illustrates the operation of an electronic device for inputting a prompt to a trained model, according to one embodiment of the present disclosure;

[0018] FIG. 5 illustrates an exemplary user interface (UI) displayed by an electronic device that generates a response to a user request, according to one embodiment of the present disclosure;

[0019] FIG. 6 illustrates a UI of an electronic device displaying a response generated using a trained model, according to one embodiment of the present disclosure;

[0020] FIG. 7 schematically illustrates a block diagram of programs executed by an electronic device according to one embodiment of the present disclosure;

[0021] FIGS. 8A and 8B illustrate UIs displayed by an electronic device according to various embodiments of the present disclosure;

[0022] FIG. 9 is a block diagram of an electronic device within a network environment according to one embodiment of the present disclosure; and

[0023] FIG. 10 is a schematic diagram of an AI system according to one embodiment of the present disclosure.

[0024] Throughout the drawings, the same drawing symbols are used to depict the same or similar elements, features, and structures.

[0025] The following description, with reference to the attached drawings, is provided to facilitate a comprehensive understanding of various embodiments of the present disclosure as defined by the claims and their equivalents. While numerous specific details are included to facilitate understanding, these are merely exemplary. Accordingly, those skilled in the art will recognize that various modifications and variations of the various embodiments described herein may be made without departing from the scope of the present disclosure. Furthermore, descriptions of well-known functions and structures may be omitted for clarity and brevity.

[0026] The terms and words used in the following description and claims are not limited to their bibliographic meanings, but are merely used by the inventors to ensure a clear and consistent understanding of the disclosure. Therefore, it should be apparent to those skilled in the art that the following description of various embodiments of the present disclosure is provided for illustrative purposes only and is not intended to limit the present disclosure, which is defined by the appended claims and their equivalents.

[0027] The term "substantially" means that a characteristic, parameter, or value need not be achieved exactly, and that deviations or variations thereof, including, for example, tolerances, measurement errors, measurement accuracy limitations, and other factors known to those skilled in the art, occur to such an extent that they do not prevent the characteristic from achieving the effect intended to be provided.

[0028] It should be recognized that the blocks and combinations of flowcharts in each flowchart can be performed by one or more computer programs containing instructions. The one or more computer programs may be entirely stored in a single memory device, or the one or more computer programs may be separated into different portions stored in different memory devices.

[0029] Any of the functions or operations disclosed herein may be processed by a single processor or a combination of processors. A single processor or a combination of processors is a circuit that performs processing and includes an application processor (AP) (e.g., a central processing unit (CPU)), a communication processor (CP) (e.g., a modem), a graphics processing unit (GPU), a neural processing unit (NPU) (e.g., an artificial intelligence (AI) chip), a Wi-Fi chip, a Bluetooth® chip, a global positioning system (GPS) chip, a nearfield communication (NFC) chip, connectivity chips, a sensor controller, a touch controller, a fingerprint sensor controller, a display driver integrated circuit (IC), an audio CODEC chip, a universal serial bus (USB) controller, a camera controller, an image processing IC, a microprocessor unit (MPU), a system on a chip (SOC), an integrated circuit, or the like.

[0030] FIG. 1 illustrates the operation of an electronic device (101) to generate a response including privacy information in response to a user request, according to one embodiment of the present disclosure.

[0031] Referring to FIG. 1, states (191, 192, 193, 194) of an electronic device (101) including a display (110) are illustrated. The hardware configuration of the electronic device (101) and / or form factors of the electronic device (101) are described with reference to FIG. 2.

[0032] According to one embodiment, an electronic device (101) may provide a function based on artificial intelligence. Artificial intelligence may be described as a technology that simulates neural activities (e.g., training activities, cognitive activities, reasoning activities, and / or creative activities) of a living organism (e.g., a human). In the present disclosure, an artificial intelligence model and / or a model may include a computational model configured to simulate neural activities based on the artificial intelligence, a software application (e.g., an agent) designed to run the computational model, hardware (e.g., a neural processing unit (NPU), a graphic processing unit (GPU), and / or a central processing unit (CPU)) configured to perform calculations represented by the computational model, or any combinations thereof.

[0033] According to one embodiment, an electronic device (101) may provide an artificial intelligence-based function using a computational model trained to output a response (e.g., text, audio, image, and / or video based on natural language) to a user request. For example, the electronic device (101) may respond to a user request and obtain at least one natural language sentence, an emoticon, or any combination thereof using the computational model. Training of the computational model may be performed (prior to installation in the electronic device (101)) based on training data, referred to as ground truth. Training of the computational model may be performed using a training algorithm, such as forward propagation and / or backward propagation. The computational model may represent a plurality of calculations tuned by a training algorithm, such as an attention mechanism, regression, a decision tree, and / or a k-nearest neighborhood.

[0034] According to one embodiment, the computational model executed by the electronic device (101) may have a structure designed to simulate neural activity. For example, the computational model may have a structure referred to as a transformer (or encoder-decoder structure). For example, the computational model may have a structure such as a convolutional model, a feedforward model, a recurrent neural network (RNN), and / or a Markov chain. The embodiment is not limited thereto, and the computational model may include a combination of computational models based on any of the structures exemplified above.

[0035] According to one embodiment, a computational model executed by the electronic device (101) may include a large language model (LLM) (or language model). The LLM may include a computational model trained based on a large amount of natural language-based information through pre-learning. The LLM may have a transformer structure trained based on an attention mechanism. The LLM may have a structure such as BERT (Bidirectional Encoder Representations from Transformer) and / or GPT (generative pre-trained transformer). The transformer structure of the computational model may include an encoder that outputs reduced-dimensional information (e.g., contextual representation) for information input to the computational model, and a decoder that outputs multi-dimensional information from the information. The encoder and the decoder may be interconnected based on a structure referred to as an attention network and / or a cross-attention network. The embodiment is not limited thereto, and the computational model may include a large vision model (LVM), and / or a large multi-modal model (LMM).

[0036] For example, a computational model executed by an electronic device (101) may be configured to process or output information (e.g., tokens) that represent the meaning of a portion of a natural language (e.g., words and / or morphemes) based on a vector space. By decoding words and / or natural language corresponding to the information output from the computational model using the vector space, the electronic device (101) may identify or obtain an output in the form of a natural language. The computational model may be trained by an algorithm such as self-supervised learning. The computational model installed in the electronic device (101) may be referred to as a trained model in terms of the trained computational model. Using the computational model, various natural language-based services, such as chatbots, translations, and / or summaries, may be provided by the electronic device (101).

[0037] In one embodiment, a trained model executed by the electronic device (101) can receive information referred to as a prompt (e.g., a set of at least one natural language sentence based on text, text in a format different from sentences (e.g., one or more words and / or phrases), an image, audio, video, or any combination thereof). While a trained model receiving a prompt based on a natural language sentence is exemplarily described, the embodiment is not limited thereto. For example, the trained model can receive a prompt including multimedia data such as an image, audio (e.g., utterances and / or music), and / or video. The electronic device (101) executing an artificial intelligence-based function can be configured to obtain a response from the trained model that satisfies the purpose and / or intent of the function using the prompt. According to one embodiment, the electronic device (101) can control the trained model using the prompt to be input to the trained model.

[0038] For example, a trained model executed by an electronic device (101) may generate responses that contain obvious errors because it has been trained to generate responses based on natural language. For example, when generating a response that includes information that was not used in training (e.g., personal information unique to the user of the electronic device (101), such as schedules, messages, profiles, contacts, locations, and / or credit information), the response output from the trained model may contain errors with respect to said information. The phenomenon of a trained model outputting a response that includes errors may be referred to as hallucination.

[0039] For example, the electronic device (101) may be designed to restrict access to a trained model to privacy information, including personal information, to enhance security. For example, the electronic device (101) may manage the trained model to access specific information based on specified rules or conditions (e.g., type of information, type of user account, type of connected network). For example, the trained model may be designed to be inaccessible to at least some privacy information. In the present disclosure, privacy information may include not only personal information but also specific categories of information that are inaccessible to the trained model. According to one embodiment, among a plurality of training models available through the electronic device (101), some training models may be managed to restrict access to (e.g., inaccessible) specific categories of information. In the above example, when the electronic device (101) controls the trained model to generate a response based on privacy information, the likelihood of hallucination occurring may increase.

[0040] For another example, the electronic device (101) may be configured to perform training on a trained model using the stored privacy information, as the privacy information is stored in the electronic device (101). In the above example, the privacy information stored after training may cause hallucination. For example, various information associated with a user or a specific account stored after training may correspond to information for processing the responses of the trained model as keywords in the present disclosure.

[0041] In one embodiment, the electronic device (101) may input a prompt to the trained model to generate a predictive (or error-free) response to information that is not accessible to the trained model in order to reduce or prevent the hallucination. The electronic device (101) may insert the personal information and / or the privacy information into the response generated from the trained model to finalize the response to be provided to the user. For example, the electronic device (101) may input a prompt to the trained model requesting a response suitable for the post-processing in order to perform post-processing on the response obtained from the trained model.

[0042] Referring to FIG. 1, states (191, 192, 193, 194) of an electronic device (101) executing a function based on a trained model are illustrated. Within state (191), the electronic device (101) may display a user interface (UI) based on a text message (e.g., a short message service (SMS)) on a display (110). The UI displayed on the display (110) of FIG. 1 may be a screen displayed by a messenger application (e.g., a messenger screen).

[0043] Referring to the state (191) of FIG. 1, the screen displayed on the display (110) may include a portion (111) (e.g., a browsing area) for displaying text messages being exchanged between the user of the electronic device (101) and another user, a portion (112) (e.g., a composing area) for composing a text message to be sent to the other user, and / or a portion (113) (e.g., a virtual keyboard area) for receiving an input for a text message to be displayed in the portion (112). Based on the state (191) of FIG. 1, the electronic device (101) may provide a conversation session in which at least one text message exchanged between the user of the electronic device (101) and the counterpart is displayed.

[0044] Within state (191) of FIG. 1, the electronic device (101) may generate or provide an artificial intelligence-based response (e.g., a text message to be transmitted to the other party). While the operation of the artificial intelligence-based electronic device (101) is described, the embodiment is not limited thereto, and the electronic device (101) may include hardware, software, or any combination thereof that performs functions corresponding to the present disclosure. For example, the electronic device (101) may identify a user request for the response. For example, the electronic device (101) may identify the user request through a messenger screen displayed on the display (110). For example, to identify the user request, the electronic device (101) may display a visual object (114). The visual object (114) in the form of an icon (e.g., including at least one of an image or text) may correspond to a function of recommending a response based on a conversation session using artificial intelligence. Based on a touch gesture (e.g., a tap gesture) and / or a mouse click on a visual object (114), the electronic device (101) may execute artificial intelligence to obtain or generate candidate responses. In one embodiment, the function of recommending responses based on a conversation session may also be executed based on an input (e.g., a voice input including a specific utterance, a specific touch input to a screen) designated to execute an intelligent assistant (e.g., Samsung® Bixby™).

[0045] Referring to FIG. 1, an electronic device (101) that receives an input for selecting a visual object (114) within a state (191) may switch to a state (192). The electronic device (101) that identifies a user request based on the visual object (114) may control a trained model to generate one or more responses suitable for the conversation session. Referring to FIG. 1, within the state (192), the electronic device (101) may display, on a display (110), a portion (121) including responses generated based on the trained model. Within the portion (121), natural language sentences reflecting situations indicated by text messages included in the conversation session may be listed. For example, in a state where text input by a user exists in a portion (112), the electronic device (101) that received the input can control a model trained using the text input in the portion (112) to generate or output one or more responses based on the text input in the portion (112) (e.g., natural language sentences including the subject and / or content of the text input in the portion (112)).

[0046] For example, in a state (191) where the other party sends a text message to the electronic device (101) to check the user's evening schedule (e.g., "What time is free tonight? Let's go eat!"), the electronic device (101) that received the input can obtain responses including natural language related to the evening schedule (e.g., "7 pm is free", "I don't have any other plans starting at 7 pm", and / or "7 pm is good for me! Are you free too?") using a trained model. An operation of the electronic device (101) generating the responses based on the trained model is described with reference to FIG. 3, FIG. 4, and / or FIG. 7.

[0047] Referring to state (192) of FIG. 1, each of the responses may have a selectable form (e.g., a button and / or a text box) within a portion (121). The electronic device (101) may receive an input for transmitting at least one of the responses to the other party via the portion (121). For example, the electronic device (101) may receive a user input for selecting a visual object (123) corresponding to one of the responses (e.g., a button containing "I have no other plans from 7 PM"). Upon receiving the user input, the electronic device (101) may switch from state (192) to state (193).

[0048] Referring to state (193) of FIG. 1, the electronic device (101) may display a visual object (131) for transmitting text (e.g., "I have no other plans from 7 p.m.") included in the visual object (123) together with the visual object (123). A visual object (131) in the form of a button containing designated text such as "Send" is illustrated as an example, but the embodiment is not limited thereto. For example, the electronic device (101) may receive an input for editing the text or displaying another screen related to the text. For example, the electronic device (101) may receive an input for at least partially modifying or removing the text, or adding an emoticon to the text. Within state (193), the operation of the electronic device (101) for receiving additional input related to the text is described with reference to FIG. 6.

[0049] Within state (193) of FIG. 1, an electronic device (101) that receives an input for selecting a visual object (131) may switch to state (194). In response to the input, the electronic device (101) may transmit text related to the input (e.g., “I have no other plans after 7 p.m.”) to the other party of the conversation session (or an external electronic device to which the other party is logged in). Within state (194), the electronic device (101) may display a visual object (141) representing the text on the display (110). The visual object (141) may have the form of a bubble containing the text. The visual object (141) may be displayed together with other text messages exchanged between the user of the electronic device (101) and the other party within the portion (111). Referring to states (192, 193), the electronic device (101) can receive an input for selecting one of a plurality of responses, and in response to the input, can transmit the response (e.g., a text message) selected by the input to the other party.

[0050] As described above, according to one embodiment, the electronic device (101) may, in response to a user request to execute artificial intelligence, generate a response based on the context in which the user request occurred (e.g., a conversation session) and privacy information about the user of the electronic device (101). The electronic device (101) may generate the response using a trained model that is inaccessible to privacy information. Since the trained model does not learn and / or access privacy information, the security of privacy information may be enhanced.

[0051] Below, with reference to FIG. 2, the structure of an electronic device (101) for driving a trained model that is inaccessible to privacy information is described.

[0052] FIG. 2 illustrates a block diagram of an electronic device (101) according to one embodiment of the present disclosure. Referring to FIG. 2, the electronic device (101) may be one of various forms of electronic devices, such as a laptop PC (personal computer) (290), smartphones (291) having various form factors (e.g., a bar-type smartphone (291-1), a foldable-type smartphone (291-2), or a sliderable (or rollable) type smartphone (291-3) described with reference to FIG. 1), a tablet PC (292), a head-mounted display (HMD) device (293), a watch (294), a cellular phone (not shown), and other similar computing devices (not shown).

[0053] In one embodiment, the electronic device (101) may be referred to as a mobile device, a user equipment (UE) (or user terminal), a multi-function device, a portable communication device, a portable device, or a server. The form factor of the electronic device (101) is not limited to the form factors illustrated in FIG. 2. For example, the electronic device (101) may be included as an electronic control unit (ECU) in a vehicle (e.g., an electric vehicle (EV)). For example, the electronic device (101) may have a form factor that is wearable by a user, such as an earbud (or wireless earphone) and / or a ring, or may have a form factor that is implantable on a body part of a user. For example, the electronic device (101) may have a form suitable for playing multimedia content.

[0054] Referring to FIG. 2, according to one embodiment, an electronic device (101) may include a processor (210) and / or a memory (220). The electronic device (101) may further include a display (110). The processor (210) may be electrically and / or operatively coupled with the memory (220) and / or the display (110). Electrical coupling of the electronic components may include a state in which a wired signal path (or a connection for wireless communication) for transmitting a signal is established between the electronic components. Operationally coupling of the electronic components may include a state in which the electronic components are directly coupled (or a state in which the electronic components are indirectly coupled) such that one of the electronic components controls another electronic component.

[0055] Referring to FIG. 2, for convenience of explanation, an electrical connection between a display (110), a processor (210), and a memory (220) is schematically illustrated. The processor (210) may be communicatively coupled to the display (110) and / or the memory (220) via one or more electronic components (e.g., a bus (202) and / or a communication bus (202)). A wired interface for transmitting information may be established between the processor (210), the memory (220), and the display (110).

[0056] The processor (210) of FIG. 2 may include circuits (e.g., processing circuits and / or cores) for performing operations on data (e.g., arithmetic operations and / or logical operations). Binary codes (e.g., instructions) representing the operations may be input to the processor (210). The processor (210) may be referred to as an application processor (AP). The processor (210) may include a central processing unit (CPU), a graphic processing unit (GPU), and / or a neural processing unit (NPU). The processor (210) may include processing circuits configured to perform functions indicated by the instructions. The number of processors (210) included in the electronic device (101) may be one or more. At least one processor included in the electronic device (101) may be configured to individually or collectively execute instructions to perform the operations of the present disclosure.

[0057] The memory (220) of FIG. 2 may include a circuit for storing data (or instructions) input to or output from the processor (210). The memory (220) may include volatile memory, such as random-access memory (RAM), and / or non-volatile memory, such as read-only memory (ROM). The non-volatile memory may be referred to as storage. The volatile memory may include, for example, at least one of dynamic RAM (DRAM), static RAM (SRAM), cache RAM, and pseudo SRAM (PSRAM). The non-volatile memory may include, for example, at least one of programmable ROM (PROM), erasable PROM (EPROM), electrically erasable PROM (EEPROM), flash memory, hard disk, compact disc, solid state drive (SSD), and embedded multimedia card (eMMC). The memory (220) may include one or more storage media (e.g., the volatile memory and / or non-volatile memory described above) distributedly located in the electronic device (101). The processor (210) of the electronic device (101) may execute instructions of the memory (220) within the electronic device (101) to perform functions and / or operations (e.g., the operations of FIG. 3) indicated by the instructions. For example, at least one processor of the electronic device (101), including the processor (210), may be configured to collectively or individually execute the instructions.

[0058] The display (110) of the electronic device (101) may include a circuit for visualizing information provided from the processor (210). The display (110) may include a liquid crystal display (LCD), a plasma display panel (PDP), and / or light emitting diodes (LEDs). The LEDs may include organic LEDs (OLEDs). The embodiment is not limited thereto, and the display (110) may include electronic paper. The display area (or active area) of the display (110) may include an area where light is emitted, formed by pixels (e.g., activated pixels) of the display (110). The display (110) may include a sensor (e.g., a touch sensor) for detecting an external object (e.g., a user's finger) on the display (110). The sensor may be included in the display (110) in the form of a panel (e.g., a touch sensor panel (TSP)). A display (110) including the above sensor may be referred to as a touch screen. The display (110) may further include a sensor (e.g., a digitizer based on electro-mechanical relays (EMR) and / or active electrostatic solution (AES)) for detecting an external object such as a stylus.

[0059] Referring to FIG. 2, programs stored in a memory (220) to be executed by a processor (210) (e.g., a (software) application (241), a response generator (242), a model (243)) and information accessible by at least one of the programs (e.g., a privacy DB (251)) are illustrated. The programs may be independently installed in the memory (220) or may be stored in the memory (220) as sub-routines (or applets or dynamic link libraries (DLLs)) of a single program.

[0060] According to one embodiment, the processor (210) of the electronic device (101) may execute an application (241) to execute a function related to artificial intelligence or provide a user experience related to the artificial intelligence. The application (241) may include not only the messenger application described with reference to FIG. 1, but also a conversational application based on voice recognition, an email, a social network service (SNS), a video streaming application, a podcast (for streaming audio), a word processor, online banking, and / or an editing application for media content (e.g., images, videos, and / or audio).

[0061] According to one embodiment, the processor (210) of the electronic device (101) may execute a model (243) to perform calculations related to artificial intelligence. The model (243) may include a trained model as described with reference to FIG. 1. While an embodiment (e.g., on-device and / or standalone) in which the calculations of the model (243) are directly performed by the processor (210) is described, the embodiment is not limited thereto. For example, the electronic device (101) may transmit a signal to an external electronic device connected via a communication circuit to control the model (243) executed on the external electronic device. For example, the operation of executing the model (243) may include not only the operation of the processor (210) directly performing the calculations of the model (243), but also the operation of communicating with a server and receiving information output from the model (243) of the server. For another example, the electronic device (101) may perform an operation to generate a summary and / or prompt of a conversation session using a model within the electronic device (101), and may communicate with an external electronic device via a communication circuit to obtain a response including an emoticon from the model within the external electronic device, or perform an operation related to a conversation session involving three or more users. For example, the electronic device (101) may use the model within the electronic device (101) when only textual content is required as a response, and may use the model within the external electronic device when content such as an image, video, or audio is required as a response. For example, the electronic device (101) may use the model within the external electronic device when judgment of a conversation, topic, or speaker related to a response is required in a multi-party conversation environment.

[0062] According to one embodiment, the processor (210) of the electronic device (101) may store information (e.g., privacy information) about the user of the electronic device (101) in the privacy DB (251) of the security area (250). Privacy information input into the electronic device (101) may be stored in the privacy DB (251). The privacy information may include personal information of the user of the electronic device (101) and information that should not be disclosed to other users (without permission).

[0063] Referring to FIG. 2, the privacy DB (251) may be stored within a secure area (250) of the memory (220). The secure area (250) may be a portion of the memory (220) that is accessible to authorized (or authenticated) users and / or programs. For example, unauthorized users and / or programs may not be permitted to read, write, and / or update the secure area (250). Information stored in the secure area (250) may be encrypted. A processor (210) executing a program that is permitted to access the secure area (250) may decrypt information (e.g., the privacy DB (251)) stored in the secure area (250).

[0064] Referring to FIG. 2, the privacy DB (251) may include DBs (e.g., a schedule DB (261), a message DB (262), a profile DB (263), a contact DB (264), a location DB (265), and / or a wallet DB (266)) for storing various information related to a user of the electronic device (101). Information about the schedule of the user of the electronic device (101) may be stored in the schedule DB (261). Information (e.g., emails, text messages, and / or voice messages) transmitted between the user of the electronic device (101) and other users may be stored in the message DB (262). Personal information (e.g., birthday, etc.) about the user of the electronic device (101) may be stored in the profile DB (263). Contact information stored by the user of the electronic device (101) may be stored in the contact DB (264). Within the location DB (265), location information stored by the user of the electronic device (101) may be stored. Within the wallet DB (266), financial information of the user of the electronic device (101) may be stored. Within the privacy DB (251), the user's privacy information may be stored in the form of vectors and / or records.

[0065] According to one embodiment, the electronic device (101) can update or manage the privacy DB (251) using contact information, schedule information, email information, messenger information, and / or SNS information stored in the memory (220). The electronic device (101) can update the privacy DB (251) using images, videos, and / or audio stored in the memory (220) (e.g., object detection, optical character recognition (OCR), and / or speech-to-text (STT)). For example, the electronic device (101) can extract card numbers and / or account numbers from images (e.g., extraction based on OCR) and store the extracted card numbers and / or account numbers as privacy information. For example, the electronic device (101) can use information obtained from images and / or videos using OCR to confirm or identify when, where, with whom, and / or what the user of the electronic device (101) did as personal information. The information stored in the privacy DB (251) may include information about a network connected to the electronic device (101) (e.g., a Wi-Fi service set identifier (SSID) and / or information for logging into the network), sensor data obtained by the electronic device (101) (e.g., a state of charge (SOC) of a battery, a location and / or moving speed indicated by a global positioning system (GPS), and / or a user's health data, activity data, and / or sensor data based on a biometric sensor such as a heart rate sensor). Information stored in the privacy DB (251) may include information obtained from external electronic devices (e.g., home appliances, wearable devices such as HMDs, and / or automobiles) (e.g., status information of the external electronic device, content displayed on the external electronic device, user activity information).

[0066] Referring to FIG. 2, an embodiment in which a privacy DB (251) is created in the memory (220) of the electronic device (101) (and the secure area (250) of the memory (220)) is described, but the embodiment is not limited thereto. The privacy DB (251) may be stored in an external electronic device different from the electronic device (101). The operation of the processor (210) accessing the privacy DB (251) may include not only the operation of accessing the privacy DB (251) stored in the secure area (250) of the memory (220), but also the operation of communicating with the external electronic device in order to access the privacy DB (251) stored in the memory of the external electronic device. For example, the processor (210) may identify the external electronic device in which the privacy DB (251) is stored based on account information logged into the electronic device (101), and may communicate with the identified external electronic device to access the privacy DB (251) stored in the external electronic device.

[0067] Referring to FIG. 2, the processor (210) may execute a response generator (242) to process a user request based on a privacy DB (251) and / or a model (243). The processor (210) may execute the response generator (242) to generate a response to a user request identified through an application (241). The processor (210) that executes the response generator (242) may generate a prompt for generating a response to the user request. The processor (210) may generate a prompt for replacing information (e.g., information stored in the privacy DB (251)) that is not accessible by the model (243) within the response with a keyword.

[0068] Referring to FIG. 2, the model (243), the application (241), and / or the response generator (242) may be stored in a general area (240) of the memory (220), which is different from the secure area (250). The model (243) executed in the general area (240) cannot access the secure area (250). The processor (210) executing the response generator (242) may access at least a portion of the secure area (250) (e.g., the privacy DB (251)) to which access is permitted based on the response generator (242). The processor (210) may execute the model (243) using a prompt generated by the response generator (242). Since access to the security area (250) based on the model (243) is not permitted, the processor (210) may execute the model (243) to generate or obtain a response that does not include information stored in the security area (250).

[0069] As described above, hallucinations may occur when a response is generated by the model (243) that involves information that was not used in training (or inaccessible information). Since the prompt input to the model (243) (e.g., a prompt generated by executing the response generator (242)) explicitly requests that the inaccessible information be expressed as a keyword (e.g., a fixed string), as described above, the information output from the model (243) may include the keyword instead of the hallucination. For example, the processor (210) may obtain information without hallucination from the model (243). Alternatively, if the processor (210) receives information from the model (243) that may cause hallucinations (e.g., text including numbers, dates, account numbers, addresses, and / or phone numbers), the processor (210) may replace and / or modify at least a portion of the information. For example, when the processor (210) identifies a type of information classified as information where hallucination may occur, it can compare it with the information in the DB (251) to determine whether they match. For example, based on the result of the comparison, the processor (210) can change at least a portion of the information using the information included in the privacy DB (251). The processor (210) can display a part where hallucination may occur differently from other parts (e.g., by applying effects such as underlining and / or italics, or by changing the font color and / or font). The processor (210) can receive information from the model (243) that emphasizes a part where hallucination may occur compared to other parts.

[0070] The processor (210), which provides a prompt to the model (243) configured to be accessible to the general area (240) of the memory (220), can obtain a first response generated according to the prompt from the model (243). The processor (210) can search for a keyword included in the first response to prevent hallucination. The processor (210), which executes the response generator (242), can replace the keyword included in the first response with information in the secure area (250) (e.g., privacy information in the privacy DB (251)). By replacing the keyword included in the first response with privacy information in the secure area (250), the processor (210) can generate a second response. The second response can be displayed on the display (110) by the processor (210), such as the responses displayed through the portion (121) in the state (192) of FIG. 1. The second response may be used as content (e.g., text) generated by the processor (210) executing the application (241) or input to the processor (210) executing the application (241).

[0071] As described above, the processor (210) may execute the response generator (242) to at least partially change the information generated by the model (243). For example, the processor (210) may change a portion of the information generated by the model (243) that should be personalized to the user of the electronic device (101). In order to prevent the natural language generated by the model (243) that has learned a large amount of natural language from including inaccurate information about the user, the processor (210) may at least partially change the natural language generated by the model (243). Since the response generated using the response generator (242) includes not only the result of recognizing the context by the model (243) but also all of the user's personal information stored in the privacy DB (251), the response desired by the user can be provided more accurately. Since the response is generated using a model (243) with restricted (e.g., blocked) access to the user's personal information (e.g., information stored in the privacy DB (251)), security issues occurring in the model (243) can also be resolved.

[0072] Below, with reference to FIG. 3, the operation of the processor (210) that executes the response generator (242) is described.

[0073] FIG. 3 illustrates a flowchart of an electronic device according to an embodiment of the present disclosure. The electronic device (101) and / or the processor (210) of FIGS. 1 and 2 may perform the operations of FIG. 3. The order in which the operations of FIG. 3 are performed is not limited to the order illustrated in FIG. 3. For example, the processor (210) of FIG. 2 may perform the operations of FIG. 3 in a different order than that illustrated in FIG. 3. For example, the processor (210) of FIG. 2 may perform at least two of the operations of FIG. 3 substantially simultaneously.

[0074] Referring to FIG. 3 , in operation (310), according to one embodiment, an electronic device may identify an event for generating a response. The event may include a user request for generating a response, or may be generated by the user request. The event may be generated when a message including text is received from an external electronic device. The event may be generated by an input related to a virtual keyboard (e.g., an input selecting a text input UI of the virtual keyboard). The event may be generated when the electronic device displays a message received (for a specified period of time). The user request of operation (310) may include an input indicating selection of the visual object (114) of FIG. 1. The user request of operation (310) may be generated to obtain content generated by a trained model (e.g., model (243) of FIG. 2 ). The user request of operation (310) may be identified by a software application running on the electronic device (e.g., application (241) of FIG. 2 ). The electronic device, having identified the user request of operation (310), may perform the remaining operations of FIG. 3 . The embodiment is not limited thereto, and the electronic device may perform the remaining operations of FIG. 3 in response to detecting an event associated with the trained model.

[0075] Referring to FIG. 3, in operation (320), according to one embodiment, an electronic device may generate a prompt. For example, the electronic device may perform operation (320) by executing the response generator (242) of FIG. 2. The prompt may include a sentence instructing generation of a response corresponding to the user request of operation (310). The prompt may include a sentence instructing the form of the response (e.g., maximum number of characters, tone, and / or format).

[0076] The prompt generated by the electronic device performing operation (320) may include a sentence intended to reduce or prevent hallucination. For example, when using a trained model that is inaccessible to privacy information, privacy information and other information may be included in the content output from the trained model (i.e., the occurrence of hallucination). The electronic device may add a sentence to the prompt indicating that a portion where privacy information is to be inserted is to be indicated by a specific keyword (or identifier, or marker, or indicator). The electronic device may add a sentence indicating that at least one response among a plurality of responses will not include privacy information.

[0077] An electronic device performing operation (320) may generate a prompt comprising information to be referenced by a trained model that is executed to generate a response. For example, an electronic device that identifies a user request to generate a response based on a text message may generate a prompt comprising text messages that have been exchanged between users (e.g., one or more text messages, as indicated by portion (111) of FIG. 1 ). In this example, the electronic device may generate a prompt comprising text messages generated within a specified time period (e.g., a period of one week from the present time), a specified number of text messages (e.g., six), and / or text messages with a specified number of characters. For example, an electronic device that identifies the user request via a first software application may generate a prompt using information associated with the first software application (e.g., text messages exchanged between users via SMS) as well as information associated with a second software application (e.g., text messages exchanged between users via SNS). To transmit large amounts of information to a trained model, the electronic device can execute a summary function based on the trained model. For example, a prompt containing a summary of text messages generated within a specified period of time can be generated. The electronic device can execute the summary function when the length and / or size of the information included in the prompt exceeds a threshold.

[0078] While examples of generating prompts based on text messages have been described, embodiments are not limited thereto. For example, an electronic device that identifies a user request to generate a response to be included in an email may generate a prompt that includes emails exchanged with the user designated as the recipient. For example, the electronic device may generate a prompt that includes text and / or images that were displayed on the display (or a UI displayed on the display) at the time the user request was identified.

[0079] Referring to FIG. 3, in operation (330), according to one embodiment, an electronic device can obtain a first response generated according to a prompt of operation (320) from a trained model (e.g., model (243) of FIG. 2). While an embodiment of controlling a model using the prompt of operation (320) is described, the embodiment is not limited thereto, and the electronic device can input information in a different format than the prompt (e.g., information described in the present disclosure and / or information self-evident from the present disclosure) into the model to obtain the first response of operation (330). The electronic device can input the prompt into the trained model. The electronic device can obtain the first response of operation (330) from the model into which the prompt has been input. The first response can include information in a form indicated by the prompt (e.g., natural language, emoticon, image, audio, and / or video). Upon receiving a prompt requesting the generation of a text message (e.g., a sentence based on natural language), the electronic device may obtain one or more sentences based on natural language from the model. The first response may include textual information in various formats, including not only sentences but also words and / or phrases.

[0080] Referring to FIG. 3, in operation (340), according to one embodiment, an electronic device may identify a keyword set by a prompt from a first response. If the electronic device generates a prompt requesting that a response be generated in the form of natural language, the information obtained from the trained model may be in the form of natural language. If the electronic device generates a prompt requesting that privacy information be replaced with a keyword, the information obtained from the trained model may include the keyword. The electronic device may search for the keyword in the first response obtained from the trained model. If the first response does not include the keyword, the electronic device may output the first response.

[0081] Referring to FIG. 3, in operation (350), according to one embodiment, an electronic device can obtain privacy information corresponding to a keyword. The electronic device can obtain privacy information of a category based on the keyword. The electronic device can search a database (e.g., privacy DB (251) of FIG. 2) stored in a secure area (e.g., security area (250) of FIG. 2) to obtain privacy information related to the keyword. The electronic device can search the database using not only the keyword but also information related to the prompt of operation (320) to obtain or identify the privacy information of operation (350). The electronic device can search the database using not only the keyword but also the remaining texts different from the keyword within the first response obtained based on operation (330).

[0082] Referring to FIG. 3, in operation (360), according to one embodiment, the electronic device may generate a second response by replacing the keywords of the first response with privacy information obtained based on operation (350). If the electronic device obtains a first response including text based on natural language, the electronic device may generate or obtain a second response including text based on natural language by replacing the keywords of the first response with privacy information of operation (350).

[0083] In one embodiment, the electronic device can identify portions of the first response to be replaced with privacy information, and change the identified portions to privacy information, even if the first response does not include keywords (or indicators) included in the prompt of operation (320). For example, if the first response includes information such as a schedule, birthday, date, amount, location, contact information, and / or account information, the information included in the first response is likely to contain errors, since the trained model cannot access privacy information. In the example, the electronic device can compare the information included in the first response with the privacy information and determine whether to change or replace the information. Depending on the specified type and / or specified format, the electronic device can extract or obtain information in the first response to be compared with the privacy information. In the example, the electronic device can change or replace the information included in the first response with privacy information stored in the secure area. The electronic device can change the information included in the first response based on whether the information included in the first response matches the privacy information in the secure area.

[0084] While one embodiment based on natural language and / or text has been described, the embodiment is not limited thereto. For example, if an electronic device identifies a user request to generate an image, the electronic device may obtain a first image using a trained model. The prompt input to the model to obtain the first image may include information (e.g., information in the form of a sentence) instructing the model to mark a portion corresponding to privacy information (e.g., the face of the user of the electronic device, text corresponding to the privacy information) with a marker (or a designated color). Upon detecting a marker (or designated color) in the first image obtained from the trained model, the electronic device may generate a second image by modifying at least a portion of the first image using the privacy information (e.g., a photo of the user's face). Even if the electronic device identifies a user request to generate video and / or audio, the electronic device may perform an operation similar to the above operation of changing the first image into a second image.

[0085] Referring to FIG. 3, in operation (370), according to one embodiment, the electronic device may output a second response. For example, the electronic device may display or output the second response on a display (e.g., a messenger screen), as in states (193, 194) of FIG. 1. The electronic device displaying the second response may visually emphasize (e.g., underline and / or adjust the font color) words and / or portions (e.g., numbers) related to privacy information within the second response.

[0086] An electronic device that outputs a second response based on operation (370) may further receive requests related to the second response. For example, the electronic device may identify or receive a transmission request for the second response. The electronic device that has identified the transmission request may transmit the second response to an external electronic device via a communication circuit.

[0087] Below, with reference to FIG. 4, the operation of the electronic device that generates the prompt of operation (320) is described.

[0088] FIG. 4 illustrates the operation of an electronic device inputting a prompt to a trained model (e.g., model (243) of FIG. 2 ), according to one embodiment of the present disclosure. The electronic device (101) and / or the processor (210) of FIGS. 1 and 2 may perform the operation of FIG. 4 . The operation of the electronic device described with reference to FIG. 4 may be related to at least one of the operations of FIG. 3 (e.g., operation (320)).

[0089] Referring to FIG. 4, a state (191) of an electronic device (101) displaying a messenger screen is illustrated. The state (191) of FIG. 4 may correspond to the state (191) of FIG. 1. The electronic device (101) may identify or receive a user request (e.g., a user request of operation (310) of FIG. 3) to generate a response in the form of a text message using a trained model (e.g., a model (243) of FIG. 2) based on an input indicating a selection of a visual object (114).

[0090] The electronic device (101) that receives the user request may generate a prompt (410). The prompt (410) may indicate a task to be performed by the model, background information (e.g., context) required to perform the task, input data to be performed on the task, the format of the result of performing the task (e.g., markdown, extended marked-up language (xml), and / or JavaScript object notation (JSON)), or any combination thereof.

[0091] Referring to FIG. 4, a portion (411) of a prompt (410) generated by an electronic device (101) may include text messages exchanged between users (e.g., a user of the electronic device (101) and a counterpart of a conversation session) as background information to be input into a model. For example, history information may be stored in the portion (411). When text messages of a conversation session linked with multiple counterparts are identified, the electronic device (101) may (optionally) input text messages of the counterparts corresponding to the last received text message into the portion (411). When text messages of a conversation session linked with multiple counterparts are identified, the electronic device (101) may display a UI for selecting at least one counterpart from the plurality of counterparts. When an input for selecting at least one counterpart is received through the UI, the electronic device (101) may input text messages related to at least one counterpart selected by the input into the portion (411).

[0092] A portion (412) of the prompt (410) may include at least one sentence indicating a task to be performed by the trained model. Referring to portion (412) of FIG. 4, the electronic device (101) may generate a prompt (410) that includes a sentence indicating what text message to output (or infer) from the perspective of a user of the electronic device (101). The embodiment is not limited thereto, and the electronic device (101) may add one or more sentences to portion (412) of the prompt (410) to indicate the tone, style, and / or language of a text message to be generated from the trained model based on text messages exchanged between users. When generating the prompt (410), the electronic device (101) may display a UI (e.g., a menu) on the display (110) for selecting the tone, style, and / or language. The embodiment is not limited thereto, and the electronic device (101) may generate a prompt (410) including at least one word provided by the user.

[0093] Referring to part (412) of FIG. 4, the electronic device (101) may generate a prompt (410) including a sentence that instructs that privacy information (e.g., personal information) be expressed as a keyword (e.g., a keyword including a designated character such as "@"). The keyword may indicate a type and / or category of privacy information to be included in the response. For example, the electronic device may generate the prompt (410) including the keyword so as not to include text that does not match information (e.g., the privacy information) that is not accessible by the trained model. The prompt (410) may include a sentence for replacing the mismatched text with the keyword (e.g., "If you don't have any access to personal data, do use "@keyword" instead of using personal information and make a reply sentence."). The prompt (410) may include a plurality of keywords (e.g., schedule, birthday, name, and / or location) that may be included in the response. Part (412) may further include information to limit the length of the response output from the trained model (e.g., the number of characters allowed by the software application).

[0094] Referring to part (413) of FIG. 4, the electronic device (101) may generate a prompt (410) indicating the format of a response to be output from a trained model. Based on the JSON format, the electronic device (101) may generate a prompt (410) including a part (413) indicating that a list of keywords included in the response should be input into a variable named "keyword" and that a response should be input into a variable named "answer." The embodiment is not limited thereto. For example, the electronic device (101) may generate a prompt (410) for outputting keywords included in the response together with the response.

[0095] Referring to FIG. 4, the electronic device (101) can input a prompt (410) into a trained model to obtain information (420) including a response (422). The information (420) can include a response (422) in the form of natural language, and information (421) indicating a keyword included in the response (422). Within the information (420), the response (422) can be set to be stored in a variable named "answer", and the information (421) indicating the keyword can be set to be stored in a variable named "keyword".

[0096] In one embodiment, the electronic device (101) may identify a keyword indicated by the prompt (410) within a response (422) included in the information (420). One or more keywords may be included (by way of example) within the prompt (410). The trained model may be trained to generate the response (422) using at least one of the keywords included in the prompt (410). For example, the prompt (410) may include a plurality of keywords (e.g., "@schedule (423)", "#account number", and / or "!address"), which are combinations of designated symbols (e.g., "@", "#", and / or "!") and words. The embodiment is not limited thereto, and the prompt (410) may include a natural language sentence including at least one keyword as an example for the response (422). For example, the electronic device (101) can search for a keyword (or word) included in the information (421) within the response (422). For example, the electronic device (101) that has obtained “schedule” as the information (421) representing the keyword, “I don’t have any other plans after dinner @schedule” as the response (422), can search for “schedule” within the response (422). The electronic device (101) can search for a combination of a designated character representing a keyword (e.g., “@”) and the keyword within the response (422) as an identifier (or delimiter) for the privacy information and / or the keyword.

[0097] Referring to FIG. 4, the electronic device (101) that has identified a keyword in the response (422) can search the DB of the electronic device (101) (e.g., the privacy DB (251) of FIG. 2) to insert privacy information into the portion of the response (422) where the keyword is located. For example, the electronic device (101) that has identified a keyword indicating a schedule ("@schedule") can search the DB of the secure area to search for privacy information to replace the keyword. The electronic device (101) can obtain privacy information by searching the DB corresponding to the keyword (e.g., the schedule DB (261) of FIG. 2). The conditions for searching the DB can be determined based on the portion (411).

[0098] For example, if a schedule is searched and there are no schedules after 7 PM, the electronic device (101) may replace the keyword in the response (422) with text representing the search result (e.g., "7 PM"). The text with the keyword changed to the search result (e.g., "I have no other schedules after 7 PM") may be displayed on the display (110).

[0099] While the state of searching for a schedule is illustrated, the embodiment is not limited thereto. For example, if a keyword such as "birthday" is included, the electronic device (101) can search the profile DB (263) of FIG. 2 and replace the keyword included in the response (422) with the privacy information, "birthday."

[0100] In one embodiment, the electronic device (101) may display a text message obtained by replacing a keyword in the response (422) with privacy information on the display (110). The embodiment is not limited thereto, and the electronic device (101) may display the response (422) before displaying the text message. The electronic device (101) may change or replace the keyword with privacy information based on an input related to the keyword included in the response (422) (e.g., a touch input for the keyword). Based on the input, the electronic device (101) may display a visual object (e.g., a pop-up window and / or a drop-down list) from which privacy information to be replaced with the keyword can be selected. Within the visual object, the electronic device (101) may display at least one candidate text to be replaced with the keyword. The candidate text may be generated or inferred from privacy information (or the user's usage pattern and / or conversation history) using a trained model. If the candidate text is not obtained, the electronic device (101) may execute another software application related to the keyword (e.g., a schedule application for an input related to a keyword corresponding to a schedule) based on the input related to the keyword. Using the other software application, the user of the electronic device (101) can (directly) confirm the information to be inputted into the keyword. Through the other software application, the electronic device (101) may obtain the text to be inputted into the keyword. For example, when the schedule application is executed and the user selects an empty time, the electronic device (101) may change or replace the keyword with the selected time.

[0101] In a state where a response (422) of FIG. 4 is displayed, the electronic device (101) may, based on the input, display a drop-down list including, as items, times (e.g., times indicated as free time by the privacy DB) that may be included in the response (422). Based on an input for selecting one of the items of the drop-down list, the electronic device (101) may change at least a portion of the response (422) into text indicating a time corresponding to the input. Using the visual object, the electronic device (101) may determine privacy information to be replaced with the keyword. The visual object may include a list of privacy information that may be inserted into a portion of the response (422) where the keyword is displayed.

[0102] Although an embodiment has been described in which the electronic device (101) inputs personal information about a user of the electronic device (101) into the response (422), the embodiment is not limited thereto. For example, the electronic device (101) may replace or change keywords in the response (422) with information about a user other than the user (e.g., a counterpart in a conversation session).

[0103] For example, if the electronic device (101) obtains a plurality of responses including a response (422), the electronic device (101) may perform an operation of searching for a keyword for each of the plurality of responses. If a specific response does not include a keyword, the electronic device (101) may display the specific response together with other responses on the display (110). The plurality of responses may be displayed in the form of a list on the display (110), such as states (192, 193, 194) of FIG. 1.

[0104] Below, the operation of an electronic device (101) that identifies a user request to execute a trained model through a UI different from the messenger screen is described.

[0105] FIG. 5 illustrates a user interface (UI) displayed by an electronic device (101) that generates a response to a user request, according to one embodiment of the present disclosure. Referring to FIG. 5, a state (501) of an electronic device (101) performing the operations of FIGS. 1 to 4 is illustrated.

[0106] Referring to state (501) of FIG. 5, the electronic device (101) may display a panel (e.g., a notification panel) on which notification messages are accumulated, on the display (110). The panel may be displayed based on a gesture performed on the display (110) (e.g., a swipe gesture performed from the top of the display (110) along the vertical direction of the display (110). The panel may display a portion (511) for adjusting setting values ​​of the electronic device (101) (e.g., brightness of the display (110), Wi-Fi connection status, Bluetooth connection status, reference direction of the screen, airplane function, and / or flash light), and a portion (512) on which notification messages are accumulated. Within the portion (512), visual objects representing software applications executed by the electronic device (101) and / or push messages transmitted to the electronic device (101) may be accumulated. Within the portion (512), the electronic device (101) may display a visual object (513) (e.g., a button including a specified text such as “clear”) for removing the visual objects accumulated in the portion (512). The electronic device (101) that receives an input related to the visual object (513) may delete or hide the visual objects included in the portion (512).

[0107] Referring to FIG. 5, an electronic device (101) that receives a text message based on SMS (or another messenger service) (e.g., “I heard it’s your birthday soon. When is it?”) may display a visual object (519) representing the text message within the panel illustrated in FIG. 5. In response to an input indicating selection of the visual object (519), the electronic device (101) may display a pop-up window (520). For example, the pop-up window (520) may be displayed as an overlay on the panel illustrated within state (501). For example, the pop-up window (520) may be displayed within a portion (512) at a location where the visual object (519) was displayed. The embodiment is not limited thereto, and the electronic device (101) that identifies an input related to the visual object (519) may switch to state (191) of FIG. 1.

[0108] An electronic device (101) displaying a pop-up window (520) may display text messages exchanged through a conversation session related to a text message corresponding to a visual object (519) within the pop-up window (520). The electronic device (101) may display a visual object (521) for generating a response based on a trained model (e.g., model (243) of FIG. 2) within the pop-up window (520), and a visual object (522) for stopping the display of the visual object (519). In response to an input for selecting the visual object (522), the electronic device (101) may stop displaying the pop-up window (520). In response to an input indicating the selection of the visual object (522), the electronic device (101) may remove or hide the visual object (519) within the portion (512).

[0109] An electronic device (101) that receives an input for selecting a visual object (521) via a pop-up window (520) may perform the operations described with reference to FIGS. 1 to 4 . For example, the electronic device (101) may generate a prompt to be input to a trained model using one or more text messages stored in a conversation session. The prompt may include a request to display privacy information as an identifier, word, and / or indicator, such as a keyword. The electronic device (101) may execute the trained model using the prompt. Since the prompt includes the one or more text messages, the one or more text messages may be input to the trained model.

[0110] The electronic device (101) may execute a trained model to obtain at least one response message from the trained model, which is to be transmitted to an external electronic device. Based on a determination that a designated identifier is included in the at least one response message, the electronic device (101) may replace the designated identifier included in the at least one response message with privacy information stored in the electronic device (101) (e.g., information stored in the privacy DB (251) of FIG. 2 ). The electronic device (101) may display the at least one response message, in which the designated identifier is replaced with the privacy information, within a pop-up window (530) displayed on the display (110). For example, the at least one response message may be displayed in a portion (531) of the pop-up window (530). While one embodiment is illustrated in which response messages including privacy information are displayed through the portion (531), the embodiment is not limited thereto. For example, the electronic device (101) may display a response message that does not (at all) contain privacy information (e.g., “Why? Give me a gift?”, “Guess?”, “Don’t you know that either?”, “I don’t know either”, “I’ll tell you later”, and / or “It’s difficult for me to tell you”), in the portion (531), together with a response message that contains privacy information.

[0111] Referring to FIG. 5, an embodiment of displaying response messages in a single language (e.g., a language specified by a user) through a portion (531) is illustrated, but the embodiment is not limited thereto. For example, the electronic device (101) may display response messages in different languages ​​through the portion (531). For example, the language of the response message may include not only the (default) language set by the user of the electronic device (101), but also the language of the text message(s) exchanged through the conversation session. The electronic device (101) may identify or confirm the language of the text message(s) exchanged through the conversation session using an artificial intelligence model. The electronic device (101) that has identified the language may add a natural language sentence to a prompt (e.g., prompt (410) of FIG. 4) indicating that a response based on the identified language will be generated. According to one embodiment, the electronic device (101) may generate and provide a single response (e.g., a natural language sentence) in multiple languages. For example, a response generated from the electronic device (101) may include a first natural language sentence in a first language and a second natural language sentence in a second language.

[0112] While displaying the pop-up window (530), the electronic device (101) may receive an input for selecting one of the response messages. For example, the electronic device (101) may identify or receive an input for selecting a visual object (539) corresponding to a specific response message (e.g., "My birthday is March 28th~ Want a gift?"). The electronic device (101) that has identified the input may display a visual object (541) for transmitting the response message corresponding to the visual object (539), such as a pop-up window (540), on the visual object (539). In response to the input indicating selection of the visual object (541), the electronic device (101) may transmit the response message to an external electronic device through a conversation session. The electronic device (101) can display a pop-up window (550) including a visual object (551) representing the response message in response to the input.

[0113] While the operation of the electronic device (101) based on text messages based on birthdays has been described, the embodiment is not limited thereto. For example, the electronic device (101) that identifies multiple topics from text messages exchanged through a conversation session can identify which of the multiple topics to generate a response message based on. For example, the electronic device (101) can, in response to an input indicating selection of a visual object (521), display a UI for selecting any one of the multiple topics identified through the conversation session before controlling the trained model. The UI can be displayed by the electronic device (101) to indicate that multiple topics have been identified. Based on the topic selected through the UI, the electronic device (101) can generate a prompt to be input into the trained model (e.g., a prompt indicating to generate a response based on the selected topic). For example, an electronic device (101) that has identified topics for each meeting location and meeting time through a conversation session may display a UI for determining which topic among the meeting location and meeting time to generate a response based on. Based on the topic selected through the UI, the electronic device (101) may generate or display one or more response messages.

[0114] The electronic device (101) may execute an artificial intelligence model (e.g., model (243) of FIG. 2) to identify one or more topics from text messages exchanged through a conversation session. When a plurality of topics are identified from the text messages exchanged through the conversation session, the electronic device (101) may obtain or generate a plurality of response messages corresponding to each of the plurality of topics using the trained model. The electronic device (101) may display a list including the obtained plurality of response messages. The embodiment is not limited thereto, and the electronic device (101) may obtain a response message based on any one of the plurality of topics. The electronic device (101) may display a topic corresponding to the response message together with the obtained response message. In response to a user input for selecting a topic different from the displayed topic, the electronic device (101) may generate or display at least one response message based on the topic selected by the user input.

[0115] While the operation of the electronic device (101) based on a conversation session performing a one-on-one chat with a single counterpart has been described, the embodiment is not limited thereto. For example, within a conversation session linked with multiple counterparts, the electronic device (101) may generate a prompt using the counterpart corresponding to the last received text message and the chat history with the counterpart. When providing response messages generated using the prompt, the electronic device (101) may receive an input (e.g., a long-press gesture to the other counterpart's text message) for generating a response message to the other counterpart. Based on the input, the electronic device (101) may generate a prompt based on the other counterpart's chat history and obtain one or more response messages from a model into which the prompt is input. For example, the prompt may (optionally) include text messages from the other counterpart and the user of the electronic device (101) among the text messages accumulated in the conversation session. Based on the number of counterparts included in the conversation session, the electronic device (101) may send a prompt to either a model within the electronic device (101) or a model within an external electronic device (e.g., a server), or request a response message. For example, for a conversation session in which a one-on-one chat is performed with one counterpart, the electronic device (101) may generate one or more response messages using the model within the electronic device (101). For example, for a conversation session linked with multiple counterparts, the electronic device (101) may request the server to transmit one or more response messages.

[0116] The embodiment is not limited thereto, and the electronic device (101) may generate a prompt indicating to generate response messages corresponding to each of the plurality of counterparts included in the conversation session. The embodiment is not limited thereto, and the electronic device (101) may display a UI for checking whether to generate a text message to be sent to a particular counterpart among the plurality of counterparts included in the conversation session. Through the UI, the electronic device (101) that receives an input for selecting a specific counterpart may generate a chat history between the counterpart and the user of the electronic device (101), and a prompt instructing to generate a text message to be sent to the counterpart. For example, the electronic device (101) that detects a plurality of counterparts, a plurality of topics, and / or a plurality of questions through the conversation session may select the counterparts, topics, and / or questions, and generate a prompt to be input to the trained model. Within the above example, the electronic device (101) that has detected multiple counterparties, multiple topics, and / or multiple questions may transmit the prompt to the server to request the server to generate a response message.

[0117] The operation of displaying a simplified UI, such as a pop-up window (520), is not limited to the operation described with reference to FIG. 5. For example, a simplified UI, such as the pop-up window (520) of FIG. 5, may be displayed on a display of a watch (e.g., the watch (294) of FIG. 2). For example, a foldable electronic device (e.g., a foldable type smartphone (291-2) of FIG. 2) may include a flexible display having a first size and a cover display having a second size smaller than the first size. For example, the foldable electronic device may display a simplified UI, such as the pop-up window (520), on the cover display, and may display the UI illustrated with reference to FIG. 1 on the flexible display. For example, a foldable electronic device may display a response message generated using a trained model, such as a pop-up window (530), within a state where the cover display is activated (e.g., in a folded state where the flexible display is not visible).

[0118] FIG. 6 illustrates a UI of an electronic device (101) displaying a response generated using a trained model (e.g., model (243) of FIG. 2), according to one embodiment of the present disclosure.

[0119] Referring to FIG. 6, a state (192) of an electronic device (101) performing the operations of FIGS. 1 to 5 is illustrated. The state (192) of FIG. 6 may correspond to the state (192) of FIG. 1. For example, the electronic device (101) may identify a user request for generating a response message using a trained model. In response to the user request, the electronic device (101) may obtain an output message from the trained model in which a portion where privacy information is to be input is replaced with a keyword. The electronic device (101) may replace the keyword with privacy information in the output message to generate a candidate response message to be displayed on the portion (121). As in the state (192) of FIG. 6, when the electronic device (101) obtains a plurality of output messages, the electronic device (101) may generate or display a plurality of candidate response messages by replacing the keyword of each of the plurality of output messages with privacy information.

[0120] Referring to FIG. 6, the electronic device (101) may display visual objects corresponding to each of the candidate response messages. Through the visual objects, the electronic device (101) may receive an input for transmitting or editing the candidate response message. For example, based on an input (e.g., a tap gesture) for a portion (610) (e.g., "7 o'clock") in which privacy information is entered within a candidate response message included in a visual object, the electronic device (101) may display a pop-up window (620) related to the privacy information, such as a state (602). The input for displaying the pop-up window (620) may be performed by the user for editing the privacy information included in the candidate response message. To guide that an input for the portion (610) can be received, the electronic device (101) may visually highlight the portion (610) relative to the rest of the candidate response message.

[0121] Referring to FIG. 6, within state (602), the electronic device (101) may execute a software application (e.g., a calendar application) related to privacy information to display a pop-up window (620). Since the portion (610) includes text related to a schedule (e.g., “7 o’clock”), the electronic device (101) may display the pop-up window (620) provided by the calendar application. Through the pop-up window (620), the user of the electronic device (101) may determine whether the text entered in the portion (610) is accurate.

[0122] Referring back to state (192) of FIG. 6, the electronic device (101) may receive an input for editing the candidate response message (e.g., an edit request for the candidate response message and / or privacy information included in the candidate response message) through the visual object (123) representing the candidate response message. For example, the electronic device (101) that detects a long-touch gesture (e.g., a gesture based on a finger in contact with the visual object (123) for a period exceeding about 1.5 seconds) on the visual object (123) may transition to state (603). Within state (603), the electronic device (101) may display the candidate response message in a text box (631) of a portion (112) of the display (110). Within state (603), the electronic device (101) may display a virtual keyboard on the portion (113). Through the above virtual keyboard, the electronic device (101) can receive input for editing a candidate response message included in a text box (631).

[0123] Within state (603) of FIG. 6, the electronic device (101) may display a visual object (632) for transmitting text (e.g., a candidate response message) displayed through a text box (631) within a portion (112) of the display (110) to the other party of the conversation session. Upon receiving an input related to the visual object (632) (e.g., a tap gesture for the visual object (632), the electronic device (101) may transmit the text displayed in the text box (631) to the other party (or the other party's external electronic device) through the conversation session.

[0124] As described above with reference to FIGS. 1 to 6, a response generated based on a trained model can be displayed through a portion (113) of a display (110) on which a virtual keyboard is displayed. For example, a program controlling the trained model (e.g., the response generator (242) of FIG. 2) can be included as part of a program for providing a virtual keyboard. Hereinafter, with reference to FIG. 7, the operation of an electronic device (101) for controlling a trained model using a program for providing a virtual keyboard will be described.

[0125] FIG. 7 schematically illustrates a block diagram of programs executed by an electronic device (101) according to one embodiment of the present disclosure. In the description of FIG. 7, descriptions that overlap with those of FIGS. 1 to 6 are omitted for convenience of explanation.

[0126] Referring to FIG. 7, a virtual keyboard (720), which is a software application for displaying a virtual keyboard as shown through a portion (113) of FIG. 1, FIG. 4, and / or FIG. 6, may be executed by the processor (210). The processor (210) may execute the virtual keyboard (720) to display a virtual keyboard including a visual object (114) of FIG. 1 and / or FIG. 4 on a display (e.g., the display (110) of FIGS. 1 to 6). For example, the processor (210) may identify or detect, through the virtual keyboard (720), a user request (e.g., a request of operation (310) of FIG. 3) to generate a text response based on a model (243).

[0127] The processor (210) that identifies the user request through the virtual keyboard (720) may generate a prompt based on the operation (320) of FIG. 3. The processor (210) may input the generated prompt into the model (243) and receive a first response from the model (243). Within the first response, the processor (210) that detects a designated keyword (or designated identifier) ​​set by the prompt may replace the designated keyword with privacy information using the privacy DB (729). The privacy DB (729) may correspond to the privacy DB (251) of FIG. 2 and may be exclusively accessed by the virtual keyboard (720). The response (728) generated using the model (243) and the privacy DB (729) may be displayed on a portion of the display (110) where the virtual keyboard (720) is displayed (e.g., portion (113) of FIG. 1 and / or FIG. 6). The processor (210) can execute an application (710) (e.g., a messenger application described with reference to FIG. 1 and / or FIG. 6) using the response (728).

[0128] As described above, according to one embodiment, the electronic device (101) can logically separate the trained model (243) and the privacy DB (729). For example, the model (243) may be restricted from accessing the privacy DB (729). The electronic device (101) may, by using a prompt input to the model (243), cause the model (243) to output a natural language that expresses a portion of the privacy DB (729) where information is to be input using reserved words (e.g., keywords, identifiers, markers, and / or indicators). The electronic device (101) may obtain a natural language to be output to the user of the electronic device (101) by replacing the reserved words with privacy information within the natural language.

[0129] FIG. 8A and FIG. 8B illustrate a UI displayed by an electronic device (101) according to various embodiments of the present disclosure.

[0130] Referring to FIGS. 8A and 8B, states (801, 802, 803, 804, 805, 806) of an electronic device (101) including a display (110) are illustrated.

[0131] Within the state (801) of FIG. 8A, the electronic device (101) can display a home screen (or launcher screen) on the display (110). Within the state (801), the electronic device (101) that has received a text message can display a pop-up object (810) including the text message on the display (110). The electronic device (101) that has received an input for a point (811) within the display (110) where the pop-up object (810) is displayed can switch from the state (801) to the state (802). The input can include a tap gesture on the point (811).

[0132] In state (802) of FIG. 8A, the electronic device (101) may execute a messenger application. In state (802), the electronic device (101) may display a messenger screen provided from the messenger application on the display (110). The messenger screen may include a portion (111) for displaying text messages exchanged through a conversation session (e.g., a conversation session related to a text message included in a pop-up object (810)) and a portion (112) for receiving a text message to be transmitted through the conversation session. The electronic device (101) may display a visual object (822) in the form of a bubble, including a text message included in the pop-up object (810), through the portion (111).

[0133] In state (802) of FIG. 8a, an electronic device (101) that receives an input for a point (823) within a display (110) where a visual object (821) (e.g., a visual object, referred to as a text box, on which one or more characters entered by a user are displayed) included in a portion (112) is displayed may switch to state (803).

[0134] Within state (803) of FIG. 8A, the electronic device (101) can display a virtual keyboard on a portion (113) of the display (110). The electronic device (101) can display response messages obtained from a trained model based on the operations described with reference to FIGS. 1 to 7 on a portion (830) of the display (110). Referring to state (803) of FIG. 8A and state (804) of FIG. 8B, visual objects (831, 832, 833) arranged on the portion (830) can correspond to response messages obtained from the trained model, respectively. The visual objects (831, 832, 833) can be scrolled within the portion (830) based on a drag gesture (e.g., a horizontal drag gesture) on the portion (830) of the display (110).

[0135] Referring to state (803) of FIG. 8A and state (804) of FIG. 8B, visual objects (831, 832, 833) arranged in portion (830) may each include text messages generated based on a text message included in a conversation session (e.g., a text message included in visual object (822) including a natural language sentence asking about evening plans). The text messages included in each of visual objects (831, 832, 833) may be generated by replacing a designated keyword representing a schedule with privacy information within responses obtained from a trained model and including the designated keyword.

[0136] Within state (803) of FIG. 8A and / or state (804) of FIG. 8B, the electronic device (101) may receive an input for selecting any one of the visual objects (831, 832, 833) located at a portion (830). For example, within state (803) of FIG. 8A, the electronic device (101) that receives an input for a point (839) within the display (110) where a visual object (831) is displayed may switch to state (805) of FIG. 8B.

[0137] Within the state (805) of FIG. 8b, the electronic device (101) can display text contained within the visual object (831) within the visual object (821). Within the state (805) of displaying text in the visual object (821), the electronic device (101) can receive an input for editing the text contained in the visual object (821) through a virtual keyboard displayed through the portion (113). Within the state (805) of FIG. 8b, the electronic device (101) that receives an input for a transmit button (850) contained in the portion (112) (e.g., an input for a point (851) within the display (110) where the transmit button (850) is displayed) can switch to the state (806) of FIG. 8b. The electronic device (101) that receives the above input can execute a function to transmit the text included in the visual object (821) to the other party (or the other party's external electronic device) through a conversation session.

[0138] Within state (806) of FIG. 8B, the electronic device (101) may transmit a text message generated by the trained model (e.g., "7 p.m. is fine"). The electronic device (101) may display a visual object (860) in the form of a bubble containing the text message within the portion (111). The visual object (860) may be positioned below the visual object (822) corresponding to the last received text message within the portion (111).

[0139] In the present disclosure, the operation of the electronic device (101) for generating text to be transmitted to a counterpart based on a messenger service (or messenger application) has been described, but the embodiments are not limited thereto. For example, when generating a comment (or post) to be added to a community (Internet bulletin board), the electronic device (101) may perform the operation of the present disclosure. For example, an electronic device (101) that has received an input for generating a product review to be registered in a community may perform the operation of the present disclosure to generate or output at least one candidate text that can be used as the product review. For example, an electronic device (101) that has received an input for generating a subtitle for a specific video (e.g., a video stored in the electronic device (101)) may perform the operation of the present disclosure to generate or obtain text to be combined with the video.

[0140] Below, the hardware configuration of the electronic device (101) of FIGS. 1 to 7 is described with reference to FIG. 9.

[0141] FIG. 9 is a block diagram of an electronic device (901) within a network environment (900) according to one embodiment of the present disclosure. Referring to FIG. 9 , in the network environment (900), the electronic device (901) may communicate with the electronic device (902) via a first network (998) (e.g., a short-range wireless communication network), or may communicate with at least one of the electronic device (904) or the server (908) via a second network (999) (e.g., a long-range wireless communication network). In one embodiment, the electronic device (901) may communicate with the electronic device (904) via the server (908). According to one embodiment, the electronic device (901) may include a processor (920), a memory (930), an input module (950), an audio output module (955), a display module (960), an audio module (970), a sensor module (976), an interface (977), a connection terminal (978), a haptic module (979), a camera module (980), a power management module (988), a battery (989), a communication module (990), a subscriber identification module (996), or an antenna module (997). In some embodiments, the electronic device (901) may omit at least one of these components (e.g., the connection terminal (978)), or may have one or more other components added. In some embodiments, some of these components (e.g., the sensor module (976), the camera module (980), or the antenna module (997)) may be integrated into one component (e.g., the display module (960)).

[0142] The processor (920) may, for example, execute software (e.g., a program (940)) to control at least one other component (e.g., a hardware or software component) of the electronic device (901) connected to the processor (920) and perform various data processing or operations. According to one embodiment, as at least a part of the data processing or operations, the processor (920) may store commands or data received from other components (e.g., a sensor module (976) or a communication module (990)) in a volatile memory (932), process the commands or data stored in the volatile memory (932), and store result data in a non-volatile memory (934). According to one embodiment, the processor (920) may include a main processor (921) (e.g., a central processing unit or an application processor) or an auxiliary processor (923) (e.g., a graphics processing unit, a neural processing unit (NPU), an image signal processor, a sensor hub processor, or a communication processor) that can operate independently or together with the main processor (921). For example, when the electronic device (901) includes the main processor (921) and the auxiliary processor (923), the auxiliary processor (923) may be configured to use less power than the main processor (921) or to be specialized for a given function. The auxiliary processor (923) may be implemented separately from the main processor (921) or as a part thereof.

[0143] The auxiliary processor (923) may control at least a portion of functions or states associated with at least one component (e.g., a display module (960), a sensor module (976), or a communication module (990)) of the electronic device (901), for example, on behalf of the main processor (921) while the main processor (921) is in an inactive (e.g., sleep) state, or together with the main processor (921) while the main processor (921) is in an active (e.g., application execution) state. In one embodiment, the auxiliary processor (923) (e.g., an image signal processor or a communication processor) may be implemented as a part of another functionally related component (e.g., a camera module (980) or a communication module (990)). In one embodiment, the auxiliary processor (923) (e.g., a neural network processing unit) may include a hardware structure specialized for processing artificial intelligence models. The artificial intelligence models may be generated through machine learning. This learning can be performed, for example, on the electronic device (901) itself where the artificial intelligence model is executed, or can be performed through a separate server (e.g., server (908)). The learning algorithm can include, for example, supervised learning, unsupervised learning, semi-supervised learning, or reinforcement learning, but is not limited to the examples described above. The artificial intelligence model can include multiple artificial neural network layers.The artificial neural network may be one of a deep neural network (DNN), a convolutional neural network (CNN), a recurrent neural network (RNN), a restricted Boltzmann machine (RBM), a deep belief network (DBN), a bidirectional recurrent deep neural network (BRDNN), a deep Q-network, or a combination of two or more of the above, but is not limited to the examples described above. In addition to, or alternatively to, a hardware structure, an artificial intelligence model may include a software structure.

[0144] The memory (930) can store various data used by at least one component (e.g., the processor (920) or the sensor module (976)) of the electronic device (901). The data can include, for example, software (e.g., the program (940)) and input data or output data for commands related thereto. The memory (930) can include a volatile memory (932) or a non-volatile memory (934).

[0145] The program (940) may be stored as software in the memory (930) and may include, for example, an operating system (942), middleware (944), or an application (946).

[0146] The input module (950) can receive commands or data to be used in a component of the electronic device (901) (e.g., a processor (920)) from an external source (e.g., a user) of the electronic device (901). The input module (950) can include, for example, a microphone, a mouse, a keyboard, a key (e.g., a button), or a digital pen (e.g., a stylus pen).

[0147] The audio output module (955) can output audio signals to the outside of the electronic device (901). The audio output module (955) can include, for example, a speaker or a receiver. The speaker can be used for general purposes, such as multimedia playback or recording playback. The receiver can be used to receive incoming calls. In one embodiment, the receiver can be implemented separately from the speaker or as part of the speaker.

[0148] The display module (960) can visually provide information to an external party (e.g., a user) of the electronic device (901). The display module (960) may include, for example, a display, a holographic device, or a projector and a control circuit for controlling the device. In one embodiment, the display module (960) may include a touch sensor configured to detect a touch, or a pressure sensor configured to measure the intensity of a force generated by the touch.

[0149] The audio module (970) can convert sound into an electrical signal, or vice versa, convert an electrical signal into sound. According to one embodiment, the audio module (970) can acquire sound through the input module (950), output sound through the sound output module (955), or an external electronic device (e.g., electronic device (902)) (e.g., speaker or headphone) directly or wirelessly connected to the electronic device (901).

[0150] The sensor module (976) can detect the operating status (e.g., power or temperature) of the electronic device (901) or the external environmental status (e.g., user status) and generate an electrical signal or data value corresponding to the detected status. According to one embodiment, the sensor module (976) can include, for example, a gesture sensor, a gyro sensor, a barometric pressure sensor, a magnetic sensor, an acceleration sensor, a grip sensor, a proximity sensor, a color sensor, an IR (infrared) sensor, a biometric sensor, a temperature sensor, a humidity sensor, or an illuminance sensor.

[0151] The interface (977) may support one or more designated protocols that may be used to directly or wirelessly connect the electronic device (901) with an external electronic device (e.g., the electronic device (902)). In one embodiment, the interface (977) may include, for example, a high definition multimedia interface (HDMI), a universal serial bus (USB) interface, an SD card interface, or an audio interface.

[0152] The connection terminal (978) may include a connector through which the electronic device (901) may be physically connected to an external electronic device (e.g., the electronic device (902)). In one embodiment, the connection terminal (978) may include, for example, an HDMI connector, a USB connector, an SD card connector, or an audio connector (e.g., a headphone connector).

[0153] The haptic module (979) can convert electrical signals into mechanical stimuli (e.g., vibration or movement) or electrical stimuli that a user can perceive through tactile or kinesthetic sensations. According to one embodiment, the haptic module (979) can include, for example, a motor, a piezoelectric element, or an electrical stimulation device.

[0154] The camera module (980) can capture still images and videos. According to one embodiment, the camera module (980) may include one or more lenses, image sensors, image signal processors, or flashes.

[0155] The power management module (988) can manage the power supplied to the electronic device (901). According to one embodiment, the power management module (988) can be implemented as, for example, at least a part of a power management integrated circuit (PMIC).

[0156] A battery (989) may power at least one component of the electronic device (901). In one embodiment, the battery (989) may include, for example, a non-rechargeable primary battery, a rechargeable secondary battery, or a fuel cell.

[0157] The communication module (990) may support the establishment of a direct (e.g., wired) communication channel or a wireless communication channel between the electronic device (901) and an external electronic device (e.g., electronic device (902), electronic device (904), or server (908)), and the performance of communication through the established communication channel. The communication module (990) may operate independently from the processor (920) (e.g., application processor) and may include one or more communication processors that support direct (e.g., wired) communication or wireless communication. According to one embodiment, the communication module (990) may include a wireless communication module (992) (e.g., a cellular communication module, a short-range wireless communication module, or a global navigation satellite system (GNSS) communication module) or a wired communication module (994) (e.g., a local area network (LAN) communication module, or a power line communication module). Any of these communication modules may communicate with an external electronic device (904) via a first network (998) (e.g., a short-range communication network such as Bluetooth, wireless fidelity (WiFi) direct, or infrared data association (IrDA)) or a second network (999) (e.g., a long-range communication network such as a legacy cellular network, a 5G network, a next-generation communication network, the Internet, or a computer network (e.g., a LAN or WAN)). These various types of communication modules may be integrated into a single component (e.g., a single chip) or implemented as multiple separate components (e.g., multiple chips). The wireless communication module (992) may use subscriber information (e.g., an international mobile subscriber identity (IMSI)) stored in the subscriber identification module (996) to verify or authenticate the electronic device (901) within a communication network such as the first network (998) or the second network (999).

[0158] The wireless communication module (992) can support 5G networks and next-generation communication technologies following the 4G network, such as NR access technology (new radio access technology). The NR access technology can support high-speed transmission of high-capacity data (eMBB (enhanced mobile broadband)), minimization of terminal power and connection of multiple terminals (mMTC (massive machine type communications)), or high reliability and low latency (URLLC (ultra-reliable and low-latency communications)). The wireless communication module (992) can support, for example, a high-frequency band (e.g., mmWave (millimeter wave) band) to achieve a high data transmission rate. The wireless communication module (992) may support various technologies for securing performance in a high-frequency band, such as beamforming, massive multiple-input and multiple-output (MIMO), full dimensional MIMO (FD-MIMO), array antenna, analog beam-forming, or large scale antenna. The wireless communication module (992) may support various requirements specified in the electronic device (901), an external electronic device (e.g., the electronic device (904)), or a network system (e.g., the second network (999)). According to one embodiment, the wireless communication module (992) can support a peak data rate (e.g., 20 Gbps or more) for eMBB realization, a loss coverage (e.g., 164 dB or less) for mMTC realization, or a U-plane latency (e.g., 0.5 ms or less for downlink (DL) and uplink (UL), or 1 ms or less for round trip) for URLLC realization.

[0159] The antenna module (997) can transmit or receive signals or power to or from an external device (e.g., an external electronic device). In one embodiment, the antenna module (997) may include an antenna including a radiator formed of a conductor or a conductive pattern formed on a substrate (e.g., a PCB). In one embodiment, the antenna module (997) may include a plurality of antennas (e.g., an array antenna). In this case, at least one antenna suitable for a communication method used in a communication network, such as the first network (998) or the second network (999), may be selected from the plurality of antennas, for example, by the communication module (990). A signal or power may be transmitted or received between the communication module (990) and the external electronic device via the at least one selected antenna. In some embodiments, in addition to the radiator, another component (e.g., a radio frequency integrated circuit (RFIC)) may be additionally formed as a part of the antenna module (997).

[0160] According to various embodiments, the antenna module (997) may form a mmWave antenna module. In one embodiment, the mmWave antenna module may include a printed circuit board, an RFIC disposed on or adjacent a first side (e.g., a bottom side) of the printed circuit board and capable of supporting a designated high frequency band (e.g., a mmWave band), and a plurality of antennas (e.g., an array antenna) disposed on or adjacent a second side (e.g., a top side or a side side) of the printed circuit board and capable of transmitting or receiving signals in the designated high frequency band.

[0161] At least some of the above components can be interconnected and exchange signals (e.g., commands or data) with each other via a communication method between peripheral devices (e.g., a bus, GPIO (general purpose input and output), SPI (serial peripheral interface), or MIPI (mobile industry processor interface)).

[0162] According to one embodiment, commands or data may be transmitted or received between the electronic device (901) and an external electronic device (904) via a server (908) connected to a second network (999). Each of the external electronic devices (902 or 904) may be the same or a different type of device as the electronic device (901). According to one embodiment, all or part of the operations executed in the electronic device (901) may be executed in one or more of the external electronic devices (902, 904, or 908). For example, when the electronic device (901) is to perform a certain function or service automatically or in response to a request from a user or another device, the electronic device (901) may, instead of or in addition to executing the function or service itself, request one or more external electronic devices to perform the function or at least a part of the service. One or more external electronic devices that receive the request may execute at least a portion of the requested function or service, or an additional function or service related to the request, and transmit the result of the execution to the electronic device (901). The electronic device (901) may process the result as is or additionally and provide it as at least a portion of a response to the request. For this purpose, cloud computing, distributed computing, mobile edge computing (MEC), or client-server computing technology may be used, for example. The electronic device (901) may provide an ultra-low latency service by using distributed computing or mobile edge computing, for example. In another embodiment, the external electronic device (904) may include an Internet of Things (IoT) device. The server (908) may be an intelligent server utilizing machine learning and / or a neural network. According to one embodiment, the external electronic device (904) or the server (908) may be included in the second network (999).The electronic device (901) can be applied to intelligent services (e.g., smart home, smart city, smart car, or healthcare) based on 5G communication technology and IoT-related technology.

[0163] FIG. 10 is a schematic diagram of an AI system according to one embodiment of the present disclosure.

[0164] Referring to FIG. 10, the AI ​​system (1000) may include an input / output interface (1010), an AI framework (1020), a generative AI model (1030), and / or a knowledge repository (e.g., a knowledge repository (1040)). The generative AI model (1030) of FIG. 10 may correspond to the model (243) of FIG. 2.

[0165] The input / output interface (1010) can receive input. The input can include user input and / or data acquired or generated by an electronic device (e.g., the electronic device (101) or the electronic device (901) described above). The data can include images, videos, and / or sensor data generated by at least one processor (e.g., at least one processor (210) or processor (920)) of the electronic device (e.g., illuminance data around the electronic device acquired from a sensor or sensor hub (e.g., a coprocessor (923)), posture data (or orientation data) of the electronic device, temperature inside the electronic device (e.g., temperature of the display (110) or temperature of the at least one processor (210)), size information of a display area of ​​the display (110), and / or images acquired through an image sensor (e.g., included in a camera module (980)) of the electronic device). The user input may include natural language, touch data obtained through touch circuitry included within the display module (960) (e.g., used to identify input from a finger and / or a stylus), images displayed (and / or to be displayed) on the display module (960), and / or video. As a non-limiting example, the user input may be received by the input / output interface (1010) together with context information. The context information may be described as additional information obtained in relation to the user input. The context information may relate to a state when the user input is received (e.g., including a state of the electronic device and / or a state surrounding the electronic device (e.g., a user state)). For example, the context information may include information about one or more software applications running within the electronic device when the user input is received.For example, the contextual information may include information about the location of the electronic device (or the location of the user of the electronic device) at the time the user input is received. For example, the user input may be integrated with the contextual information. For example, the user input integrated with the contextual information may be received by the input / output interface (1010).

[0166] The input / output interface (1010) can transmit (or provide) output. The output may include a result (or result information) generated or acquired by the AI ​​system (1000) based at least in part on the input. The format of the output may vary. For example, the output may include natural language. For example, the output may include content (e.g., including media content and / or multimedia content). For example, the output may include an action related to a user of the electronic device. For example, the output may have a format according to a user setting of the electronic device.

[0167] The input / output interface (1010) can be described as a user query / response interface (1010).

[0168] The AI ​​framework (1020) can be used to obtain information (or data) about the input from the input / output interface (1010) and control one or more components related to the AI ​​system (1000) using the obtained information.

[0169] For example, the prompt design component (1021) within the AI ​​framework (1020) can use the acquired information to generate or obtain prompts for a generative AI model (1030) (e.g., including a large language model (LLM) or a large multimodal model (LMM)). For example, the prompt design component (1021) can be described as an AI component that utilizes a learning algorithm and / or a neural network to provide enhanced prompts over time. For example, the prompt design component (1021) can use the acquired information to access a knowledge component (e.g., a knowledge repository (1040)) that includes user preference data, a prompt library, and / or prompt examples to generate or obtain prompts. The generated prompts can be provided to the generative AI model (1030) (e.g., including an LLM or LMM).

[0170] For example, the API / plugin management component (1022) within the AI ​​framework (1020) may be utilized to support communication for additional information requested (or induced) in connection with the prompt provided (or to be provided) to the generative AI model (1030). For example, the API / plugin management component (1022) may be utilized to create or establish channels for communication with various data sources (e.g., knowledge repositories (1040)). For example, the API / plugin management component (1022) may support access to at least some of the data sources. For example, the API / plugin management component (1022) may be utilized to request another component (e.g., an application / service component (1050)) to perform feedback (or response) according to the prompt. As a non-limiting example, information obtained (or generated) through the API / plugin management component (1022) may be provided to the prompt design component (1021) for generating a prompt. As a non-limiting example, information obtained (or generated) through the API / plugin management component (1022) may be provided to the generative AI model (1030).

[0171] For example, an enhancement component (e.g., a refinement component (1023)) within the AI ​​framework (1020) can at least partially tune (or adjust) (or change) the results (e.g., content) obtained (or output) from the generative AI model (1030). For example, the enhancement component (e.g., a refinement component (1023)) can determine or verify whether the content obtained from the generative AI model (1030) is relevant to the input. For example, the enhancement component (e.g., a refinement component (1023)) can determine or verify whether the content obtained from the generative AI model (1030) contains biased content. For example, the enhancement component (e.g., a refinement component (1023)) can determine or verify whether the content obtained from the generative AI model (1030) contains harmful content. For example, an enhancement component (e.g., a refinement component (1023)) may support or assist in performing additional processing to improve content obtained from a generative AI model (1030). For example, an enhancement component (e.g., a refinement component (1023)) may support providing hints to a user to improve the content.

[0172] A generative AI model (1030) can be described as an artificial intelligence neural network that generates feedback in response to a prompt. For example, the feedback may include additional data and / or information related to the prompt, but relative to the prompt. For example, the feedback may include new content related to the prompt. For example, the generative AI model (1030) may include a model that generates images and / or a model that generates language. For example, the model that generates images may include a generative adversarial network (GAN) and / or a variational autoencoder (VAE). For example, the model that generates images may include a diffusion-based generative model (e.g., a transformer VAE). For example, the model that generates language may include CHAT-GPT 3 and / or CHAT-GPT 4. For example, a generative AI model (1030) may include an LMM that generates the feedback by recognizing text, images, and / or speech.

[0173] As a non-limiting example, the AI ​​framework (1020) and / or the generative AI model (1030) may be included within an AI module (e.g., including a processing circuit) within the electronic device. For example, the AI ​​module may be operatively coupled with at least one processor (e.g., at least one processor (210) or processor (920)) of the electronic device. For example, the AI ​​module may be operatively coupled with a display driving circuit (e.g., a display driving circuit) of the electronic device. For example, the AI ​​module may be operatively coupled with a sensor hub of the electronic device for one or more sensors within the electronic device.

[0174] In one embodiment, a method may be required to block access to privacy information by an automated agent, such as an artificial intelligence model. In one embodiment, a method may be required to obtain information including privacy information using a trained model that is inaccessible to privacy information. In one embodiment, a method may be required to add the privacy information to output information obtained from the trained model that is inaccessible to privacy information based on post-processing. As described above, according to one embodiment, an electronic device (e.g., the electronic device 101 of FIG. 1 and / or the electronic device 901 of FIG. 9) may include one or more storage media, a memory (e.g., the memory 220 of FIG. 2) that stores instructions, and at least one processor (e.g., the processor 210 of FIG. 2) that includes a processing circuit. The instructions, when individually or collectively executed by the at least one processor, may cause the electronic device to identify an event for generating a response. The instructions, when individually or collectively executed by the at least one processor, may cause the electronic device to generate a prompt (e.g., prompt (410) of FIG. 4) to replace information associated with a second storage area (e.g., secure area (250) of FIG. 2) different from a first storage area (e.g., general area (240) of FIG. 2) of the memory with a keyword based on identifying the event. The instructions, when individually or collectively executed by the at least one processor, may cause the electronic device to provide the prompt to a trained model within the electronic device configured to access the first storage area of ​​the memory (e.g., general area (240) of FIG. 2) or the second storage area.The instructions, when individually or collectively executed by the at least one processor, may cause the electronic device to obtain a first response generated according to the prompt from the trained model. The instructions, when individually or collectively executed by the at least one processor, may cause the electronic device to generate a second response by replacing the keyword included in the first response with privacy information in the second storage area of ​​the memory (e.g., the secure area (250) of FIG. 2 ). In one embodiment, the electronic device may block access to privacy information by an automated agent, such as an artificial intelligence model. In one embodiment, the electronic device may obtain information including privacy information using a trained model that is inaccessible to privacy information. In one embodiment, the electronic device may add the privacy information to output information obtained from a trained model that is inaccessible to privacy information based on postprocessing.

[0175] For example, the electronic device may include a display (e.g., display (110) of FIG. 2). The instructions, when individually or collectively executed by the at least one processor, may cause the electronic device to identify a user request via a messenger screen displayed on the display. The instructions, when individually or collectively executed by the at least one processor, may cause the electronic device to generate the prompt, the prompt including a plurality of text messages, which may be displayed within the messenger screen.

[0176] For example, the electronic device may include a communication circuit. The instructions, when individually or collectively executed by the at least one processor, may cause the electronic device to display the second response on the display. The instructions, when individually or collectively executed by the at least one processor, may cause the electronic device to transmit the second response to an external electronic device through the communication circuit based on identifying a request to transmit the second response.

[0177] For example, the instructions, when individually or collectively executed by the at least one processor, may cause the electronic device to identify, through the second response displayed on the display, a request to edit the privacy information included in the second response. The instructions, when individually or collectively executed by the at least one processor, may cause the electronic device to display, on the display, a pop-up window of a software application related to the privacy information, superimposed on the messenger screen, based on identifying the edit request.

[0178] For example, the instructions, when individually or collectively executed by the at least one processor, may cause the electronic device to retrieve the privacy information contained in at least one of a schedule database, a messenger database, a contact database, or a wallet database related to a user of the electronic device, and generate the second response.

[0179] For example, the instructions, when individually or collectively executed by the at least one processor, may cause the electronic device to generate the prompt including the keyword such that the first response does not include text that differs from information stored in the schedule database.

[0180] For example, the instructions, when individually or collectively executed by the at least one processor, may cause the electronic device to generate the prompt for outputting, together with the first response, a list of keywords included in the first response.

[0181] According to one embodiment, an electronic device as described above may include at least one processor, the processor including one or more storage media, a memory storing instructions, and a processing circuit. The instructions, when individually or collectively executed by the at least one processor, may cause the electronic device to provide a conversation session in which at least one message exchanged between a first user of the electronic device and a second user of another electronic device is displayed. The instructions, when individually or collectively executed by the at least one processor, may cause the electronic device to, in response to a first request, input the at least one message into a first model, the first model being configured to generate a string based on a provided input, and to obtain a first message for transmission to the other electronic device from the first model. The instructions, when individually or collectively executed by the at least one processor, may cause the electronic device to display a second message, wherein the designated identifier is replaced with privacy information stored on the electronic device, at least in part based on a determination that the first message includes a designated identifier. The instructions, when individually or collectively executed by the at least one processor, may cause the electronic device to transmit the second message to the other electronic device through the conversation session, in response to a second request.

[0182] For example, the specified identifier may correspond to any one of a plurality of identifiers provided to the first model along with the at least one message.

[0183] For example, the specified identifier may include a combination of text that will be changed with respect to the privacy information, and symbols that will not be changed with respect to the privacy information.

[0184] For example, the first message may include at least one sentence containing a portion corresponding to the designated identifier. The instructions, when individually or collectively executed by the at least one processor, may cause the electronic device to identify a type of the privacy information using the designated identifier. The instructions, when individually or collectively executed by the at least one processor, may cause the electronic device to identify the privacy information using the identified type and the remaining portion of the at least one sentence.

[0185] For example, the instructions, when individually or collectively executed by the at least one processor, may cause the electronic device to display the first message before the second message is displayed. The instructions, when individually or collectively executed by the at least one processor, may cause the electronic device to change the designated identifier included in the first message to the privacy information based on an input for the designated identifier.

[0186] For example, the instructions, when individually or collectively executed by the at least one processor, may cause the electronic device, in response to the input, to display a visual object from which the first user can select the privacy information corresponding to the designated identifier of the first message.

[0187] For example, the privacy information may include information about the second user.

[0188] For example, the instructions, when individually or collectively executed by the at least one processor, may cause the electronic device to generate a summarized message to be input into the first model by summarizing the at least one message using a second model that is different from the first model, based on the at least one message being longer than a specified length.

[0189] For example, the instructions, when individually or collectively executed by the at least one processor, may cause the electronic device to obtain a third message that does not include the designated identifier. The instructions, when individually or collectively executed by the at least one processor, may cause the electronic device to display the third message together with the second message.

[0190] For example, the second request may include an input for selecting the second message from among the second message and the third message.

[0191] In one embodiment, as described above, a method of an electronic device may be provided. The method may include providing a conversation session in which at least one message exchanged between a first user of the electronic device and a second user of another electronic device is displayed. The method may include, in response to a first request, inputting the at least one message into a first model configured to generate a string based on provided input, thereby obtaining a first message for transmission to the other electronic device from the first model. The method may include, based at least in part on a determination that the first message includes a designated identifier, displaying a second message in which the designated identifier is replaced with privacy information stored in the electronic device. The method may include, in response to a second request, transmitting the second message to the other electronic device through the conversation session.

[0192] For example, the specified identifier may correspond to any one of a plurality of identifiers provided to the first model along with the at least one message.

[0193] For example, the specified identifier may include a combination of text that will be changed with respect to the privacy information, and symbols that will not be changed with respect to the privacy information.

[0194] For example, the first message may include at least one sentence containing a portion corresponding to the designated identifier. The method may include an operation of identifying a type of the privacy information using the designated identifier. The method may include an operation of identifying the privacy information using the identified type and the remaining portion of the at least one sentence.

[0195] For example, the method may include an action of displaying the first message before the second message is displayed. The action of displaying the first message may include an action of changing the designated identifier to the privacy information based on an input for the designated identifier.

[0196] For example, the act of displaying the first message may include, in response to the input, displaying a visual object from which the first user can select the privacy information corresponding to the designated identifier of the first message.

[0197] For example, the privacy information may include information about the second user.

[0198] For example, the operation of obtaining the first message may include an operation of generating a summarized message to be input into the first model by summarizing the at least one message using a second model different from the first model, based on the at least one message longer than a specified length.

[0199] For example, the action of obtaining the first message may include the action of obtaining a third message that does not include the specified identifier. The action of displaying the second message may include the action of displaying the third message together with the second message.

[0200] For example, the second request may include an input for selecting the second message from among the second message and the third message.

[0201] In one embodiment, as described above, a non-transitory computer-readable storage medium storing instructions may be provided. The instructions, when executed by an electronic device including a memory, may cause the electronic device to identify a user request. The instructions, when executed by the electronic device, may cause the electronic device to generate a response to the user request and to generate a prompt for replacing inaccessible information to be included in the response with a keyword. The instructions, when executed by the electronic device, may cause the electronic device to provide the prompt to a trained model within the electronic device, the model being configured to access a first storage area of ​​the memory. The instructions, when executed by the electronic device, may cause the electronic device to obtain a first response generated in response to the prompt from the trained model. The instructions, when executed by the electronic device, may cause the electronic device to generate a second response by replacing the keyword included in the first response with privacy information in a second storage area of ​​the memory.

[0202] For example, the instructions, when executed by the electronic device including the display, may cause the electronic device to identify a user request through a messenger screen displayed on the display. The instructions, when executed by the electronic device, may cause the electronic device to generate the prompt, the prompt including a plurality of text messages, which may be displayed within the messenger screen.

[0203] For example, the instructions, when executed by the electronic device including the communication circuit, may cause the electronic device to display the second response within the messenger screen. The instructions, when executed by the electronic device, may cause the electronic device to transmit the second response to an external electronic device via the communication circuit based on identifying a request to transmit the second response.

[0204] In one embodiment, a non-transitory computer-readable storage medium storing instructions may be provided. The instructions, when executed by an electronic device including a memory, may cause the electronic device to perform operations. The operations may include providing a chat session in which at least one message is displayed to be exchanged between a first user of the electronic device and a second user of another electronic device; inputting the at least one message into a first model configured to generate character strings based on provided input, thereby obtaining a first message to be transmitted from the first model to the other electronic device in response to a first request; displaying a second message in which the designated identifier is replaced with privacy information stored in the electronic device, based at least on a determination that the first message includes a designated identifier; and transmitting the second message to the other electronic device through the chat session based on a second request.

[0205] For example, the first model may include a model trained to have limited access to privacy information, wherein the privacy information may include personal information unique to the first user.

[0206] Electronic devices according to the various embodiments disclosed in this document may take various forms. Electronic devices may include, for example, portable communication devices (e.g., smartphones), computer devices, portable multimedia devices, portable medical devices, cameras, wearable devices, or home appliances. Electronic devices according to the embodiments of this document are not limited to the aforementioned devices.

[0207] The various embodiments of this document and the terminology used herein are not intended to limit the technical features described in this document to specific embodiments, but should be understood to include various modifications, equivalents, or substitutes of the embodiments. In this document, phrases such as "A or B," "at least one of A and B," "at least one of A or B," "A, B, or C," "at least one of A, B, and C," and "at least one of A, B, or C" can each include any one of the items listed together in the corresponding phrase among the phrases, or all possible combinations thereof. Terms such as "first," "second," or "first" or "second" may be used simply to distinguish the corresponding element from other corresponding elements and do not limit the corresponding elements in any other respect (e.g., importance or order). When a component (e.g., a first component) is referred to as being “coupled” or “connected” to another component (e.g., a second component), with or without the terms “functionally” or “communicatively,” it means that the component can be connected to the other component directly (e.g., wired), wirelessly, or through a third component.

[0208] The term "module" used in various embodiments of this document may include a unit implemented in hardware, software, or firmware, and may be used interchangeably with terms such as logic, logic block, component, or circuit. A module may be an integral component, or a minimum unit or part of such a component that performs one or more functions. For example, according to one embodiment, a module may be implemented in the form of an application-specific integrated circuit (ASIC).

[0209] Various embodiments of the present document may be implemented as software (e.g., a program (940)) including one or more instructions stored in a storage medium (e.g., an internal memory (936) or an external memory (938)) readable by a machine (e.g., an electronic device (901)). For example, a processor (e.g., a processor (920)) of the machine (e.g., an electronic device (901)) may call at least one instruction among the one or more instructions stored from the storage medium and execute it. This enables the machine to operate to perform at least one function according to the at least one called instruction. The one or more instructions may include code generated by a compiler or code executable by an interpreter. The machine-readable storage medium may be provided in the form of a non-transitory storage medium. Here, 'non-transitory' simply means that the storage medium is a tangible device and does not contain signals (e.g., electromagnetic waves), and the term does not distinguish between cases where data is stored semi-permanently or temporarily on the storage medium.

[0210] According to one embodiment, the method according to various embodiments disclosed in this document may be provided as included in a computer program product. The computer program product may be traded as a commodity between a seller and a buyer. The computer program product may be distributed in the form of a machine-readable storage medium (e.g., compact disc read-only memory (CD-ROM)), or may be distributed online (e.g., downloaded or uploaded) via an application store (e.g., Play Store™) or directly between two user devices (e.g., smart phones). In the case of online distribution, at least a portion of the computer program product may be temporarily stored or temporarily generated in a machine-readable storage medium, such as the memory of a manufacturer's server, an application store's server, or an intermediary server.

[0211] According to various embodiments, each component (e.g., a module or a program) of the above-described components may include one or more entities, and some of the entities may be separated and placed in other components. According to various embodiments, one or more components or operations of the aforementioned components may be omitted, or one or more other components or operations may be added. Alternatively or additionally, a plurality of components (e.g., a module or a program) may be integrated into a single component. In such a case, the integrated component may perform one or more functions of each of the plurality of components identically or similarly to those performed by the corresponding component among the plurality of components prior to the integration. According to various embodiments, the operations performed by a module, program, or other component may be executed sequentially, in parallel, iteratively, or heuristically, or one or more of the operations may be executed in a different order, omitted, or one or more other operations may be added.

[0212] As used herein, the term "if" will be understood to mean "when, upon," "in response to determining," or "in response to detecting," depending on the context. Similarly, "if it is determined to," or "if [the stated condition or event] is detected," will optionally be understood to mean "upon determining," or "in response to determining," "upon detecting [the stated condition or event]," or "in response to detecting [the stated condition or event]."

[0213] The devices described above may be implemented as hardware components, software components, and / or a combination of hardware components and software components. For example, the devices and components described in the embodiments may be implemented using one or more general-purpose computers or special-purpose computers, such as a processor, a controller, an arithmetic logic unit (ALU), a digital signal processor, a microcomputer, a field programmable gate array (FPGA), a programmable logic unit (PLU), a microprocessor, or any other device capable of executing instructions and responding to them. The processing device may execute an operating system (OS) and one or more software applications running on the operating system. The processing device may also access, store, manipulate, process, and generate data in response to the execution of the software. For ease of understanding, the processing device is sometimes described as being used alone; however, one of ordinary skill in the art will recognize that the processing device may include multiple processing elements and / or multiple types of processing elements. For example, a processing unit may include multiple processors, or a processor and a controller. Other processing configurations, such as parallel processors, are also possible.

[0214] Software may include a computer program, code, instructions, or a combination of one or more of these, which may configure a processing device to perform a desired operation or may independently or collectively command the processing device. The software and / or data may be embodied in any type of machine, component, physical device, computer storage medium, or device for interpretation by the processing device or for providing instructions or data to the processing device. The software may also be distributed over networked computer systems and stored or executed in a distributed manner. The software and data may be stored on one or more computer-readable recording media.

[0215] The method according to the embodiment may be implemented in the form of program commands that can be executed through various computer means and recorded on a computer-readable medium. In this case, the medium may be one that continuously stores a computer-executable program or one that temporarily stores it for execution or download. In addition, the medium may be various recording means or storage means in the form of a single or multiple hardware combinations, and is not limited to a medium directly connected to a computer system, but may also be distributed over a network. Examples of the medium may include magnetic media such as hard disks, floppy disks, and magnetic tapes, optical recording media such as CD-ROMs and DVDs, magneto-optical media such as floptical disks, and those configured to store program commands, including ROM, RAM, and flash memory. In addition, examples of other media may include recording media or storage media managed by app stores that distribute applications, sites that supply or distribute various software, servers, etc.

[0216] Although the embodiments described above have been described by way of limited examples and drawings, those skilled in the art will appreciate that various modifications and variations can be made based on the above teachings. For example, appropriate results can still be achieved even if the described techniques are performed in a different order than described, and / or components of the described systems, structures, devices, circuits, etc. are combined or combined in a different manner than described, or are replaced or substituted with other components or equivalents.

[0217] It should be understood that the various embodiments of the present disclosure, as described in the claims and specification, may be implemented in the form of hardware, software, or a combination of hardware and software.

[0218] Any software may be stored on a non-transitory computer-readable storage medium. The non-transitory computer-readable storage medium stores one or more computer programs (software modules), and the one or more computer programs include computer-executable instructions that, when individually or collectively executed by one or more processors of an electronic device, cause the electronic device to perform the method of the present disclosure.

[0219] Any software may be stored in a form of volatile or non-volatile storage, such as, for example, a storage device such as a read only memory (ROM), which may be removable, rewritable, or otherwise, or a form of memory such as a random access memory (RAM), memory chips, devices or integrated circuits, or an optical or magnetically readable medium such as, for example, a compact disk (CD), a digital versatile disk (DVD), a magnetic disk, or a magnetic tape, or the like. It will be appreciated that the storage devices and storage media are various embodiments of non-transitory machine-readable storage media suitable for storing a computer program or computer programs that, when executed, include instructions that implement various embodiments of the present disclosure. Accordingly, various embodiments provide a program comprising code for implementing an apparatus or method as claimed in any of the claims of the specification, and a non-transitory machine-readable storage medium storing such a program and the like.

[0220] While the present disclosure has been illustrated and described with reference to various embodiments, it will be apparent to those skilled in the art that various changes in form and detail may be made therein without departing from the scope of the present disclosure as defined by the appended claims and their equivalents.

Claims

1. In electronic devices, A memory including one or more storage media for storing instructions; and At least one processor comprising a processing circuit, wherein the at least one processor is communicatively coupled to the memory, The above instructions, when individually or collectively executed by the at least one processor, cause the electronic device to: Identify the event for generating a response; Based on identifying the above event, generating a prompt to replace information associated with a second storage area different from the first storage area of ​​the memory with a keyword; Providing the prompt to a trained model within the electronic device configured to access the first storage area among the first storage area or the second storage area of ​​the memory; Obtaining a first response generated according to the prompt from the trained model; and Causing a second response to be generated by replacing the keyword included in the first response with privacy information in the second storage area of ​​the memory. Electronic devices.

2. In claim 1, Including more displays, The above instructions, when individually or collectively executed by the at least one processor, cause the electronic device to: Identifying a user request through a messenger screen displayed on the above display; and further causing the prompt to be generated, comprising a plurality of text messages that can be displayed within the messenger screen; Electronic devices.

3. In claim 2, Further including communication circuits, The above instructions, when individually or collectively executed by the at least one processor, cause the electronic device to: Displaying the second response on the display; and Further causing the second response to be transmitted to an external electronic device through the communication circuit based on identifying a transmission request for the second response. Electronic devices.

4. In claim 3, the instructions, when individually or collectively executed by the at least one processor, cause the electronic device to: Through the second response displayed on the display, identify a request to edit the privacy information included in the second response; Further causing a pop-up window of a software application related to the privacy information to be displayed on the display, overlaid on the messenger screen, based on identifying the above editing request. Electronic devices.

5. In claims 1 to 4, the instructions, when individually or collectively executed by the at least one processor, cause the electronic device to: In relation to the user of the above electronic device: Schedule database; Messenger Database; Contact database; or Wallet database, further causing the second response to be generated by retrieving the privacy information contained in at least one of Electronic devices.

6. In claim 5, the instructions, when individually or collectively executed by the at least one processor, cause the electronic device to: further causing the prompt to be generated including the keyword so that the first response does not contain text different from the information stored in the schedule database. Electronic devices.

7. In claims 1 to 6, the instructions, when individually or collectively executed by the at least one processor, cause the electronic device to: Further causing the prompt to be generated, together with the first response, to output a list of keywords included in the first response, Electronic devices.

8. In electronic devices, A memory including one or more storage media for storing instructions; and At least one processor comprising a processing circuit, wherein the at least one processor is communicatively coupled to the memory, The above instructions, when individually or collectively executed by the at least one processor, cause the electronic device to: Providing a conversation session in which at least one message is displayed, exchanged between a first user of the electronic device and a second user of another electronic device; In response to the first request, inputting at least one message into the first model, the first model being configured to generate a string based on the provided input, and obtaining a first message for transmission to the other electronic device from the first model; Displaying a second message, wherein the specified identifier is replaced with privacy information stored on the electronic device, based at least in part on a determination that the first message contains a specified identifier; and In response to the second request, causing said second message to be transmitted to said other electronic device through said conversation session, Electronic devices.

9. In claim 8, the designated identifier is: corresponding to any one of the plurality of identifiers provided to the first model together with at least one message, Electronic devices.

10. In claims 8 to 9, the designated identifier is: A combination of text that will be changed with respect to the above privacy information, and symbols that will not be changed with respect to the above privacy information, Electronic devices.

11. In claims 8 to 10, the first message is: Contains at least one sentence containing a portion corresponding to the above-mentioned identifier; and The above instructions, when individually or collectively executed by the at least one processor, cause the electronic device to: Using the above-mentioned identifier, identifying the type of the privacy information; and Further causing the privacy information to be identified by using the identified type and the remaining part of the at least one sentence, Electronic devices.

12. In claims 8 to 11, the instructions, when individually or collectively executed by the at least one processor, cause the electronic device to: Before the second message is displayed, the first message is displayed; Further causing the designated identifier to be changed to the privacy information based on the input for the designated identifier included in the first message. Electronic devices.

13. In claim 12, the instructions, when individually or collectively executed by the at least one processor, cause the electronic device to: In response to said input, further causing said first user to display a visual object from which said privacy information corresponding to said designated identifier of said first message can be selected, Electronic devices.

14. In claims 8 to 13, the privacy information is: Containing information about the second user, Electronic devices.

15. In claims 8 to 14, the instructions, when individually or collectively executed by the at least one processor, cause the electronic device to: Further causing the at least one message to be summarized using a second model different from the first model based on the at least one message longer than the specified length, to generate a summarized message to be input to the first model. Electronic devices.

Images