# US 2025/0158940 A1 · Systems and methods for generating contextual replies to media content using artificial intelligence

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2025/0158940 A1
- Document ID: US20250158940A1
- Status: Published application
- Kind: A1
- Application number: 18/505,875
- Filed: 2023-11-09
- Published: 2025-05-15
- Earliest claimed: 2023-11-09
- Assignees: Snap Inc.
- Applicants: Snap Inc.
- Inventors: Dennis Jin
- Web page: https://patentgrep.com/patent/US20250158940A1

## CPC

- H04L 51/02: using automatic reactions or user delegation, e.g. automatic replies or chatbot-generated messages
- G06F 3/0237: {using prediction or retrieval techniques}
- G06F 16/90332: {Natural language query formulation or dialogue systems}
- G06F 40/166: Editing, e.g. inserting or deleting
- H04L 51/10: Multimedia information
- H04L 51/52: for supporting social networking services
- H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M)
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

A system and method for generating contextually relevant reply suggestions for media posts is disclosed. The system analyzes media content to identify visual objects, scenes, text, and metadata attributes using computer vision techniques. Identified objects and attributes are incorporated into structured prompt templates to construct detailed natural language descriptions of the media context. The prompts are provided to a text generation artificial intelligence (AI) that outputs a plurality of contextual reply suggestions based on the media analysis. Suggestions are displayed as selectable options adjacent to the media post. Users can cycle through suggestions and select a reply to send. Selections are logged to improve the AI model. Feedback on suggestion quality can also be collected. By integrating computer vision and AI generation driven by engineered prompts, the system produces highly relevant, personalized responses tailored to media content. The techniques enhance user engagement with media posts through intelligent AI reply suggestions.

## Claims

### Claim 1

A method comprising:
causing display of a presentation of media content at a client device, the media content comprising media attributes;
receiving a request to generate a reply to the media content;
accessing contextual information associated with the client device responsive to the request to generate the reply to the media content;
generating a plurality of reply suggestions based on the media attributes and the contextual information; and
causing display of the plurality of reply suggestions at the client device.

### Claim 2 (depends on 1)

The method of claim 1, wherein the receiving the request to generate the reply includes:
receiving an input that selects a graphical icon; and
generating the request to generate the reply based on the input.

### Claim 3 (depends on 1)

The method of claim 1, wherein the media content comprises a depiction of one or more objects, and wherein the generating the plurality of reply suggestions includes:
identifying the one or more objects based on the depiction of the one or more objects and the media attributes; and
generating the plurality of reply suggestions based on the one or more objects and the contextual information.

### Claim 4 (depends on 3)

The method of claim 3, wherein the identifying the one or more objects includes:
performing object recognition upon the media content; and
identifying the one or more objects based on the object recognition.

### Claim 5 (depends on 1)

The method of claim 1, wherein the generating the plurality of reply suggestions includes:
generating a prompt based on a predefined prompt template, the media attributes, and the contextual information;
providing the prompt to an artificial intelligence (AI) system; and
receiving, from the AI system, the plurality of reply suggestions.

### Claim 6 (depends on 5)

The method of claim 5, further comprising:
selecting the predefined prompt template from among a plurality of prompt templates based on one or more of the contextual information and the media attributes of the media content.

### Claim 7 (depends on 1)

The method of claim 1, wherein the causing display of the plurality of reply suggestions at the client device includes:
causing display of a first reply suggestion within a text input field at the client device;
receiving an input that comprises an input attribute;
responsive to the input, and based on the input attribute, causing display of a second reply suggestion within the text input field.

### Claim 8 (depends on 1)

The method of claim 1, wherein the media content comprises image data.

### Claim 9 (depends on 1)

The method of claim 1, wherein the contextual information includes one or more of:
location data;
temporal data; and
user profile data.

### Claim 10

A system comprising:
one or more processors; and
a memory comprising instructions which, when executed by the one or more processors, cause the one or more processors to perform operations comprising:
causing display of a presentation of media content at a client device, the media content comprising media attributes;
receiving a request to generate a reply to the media content;
accessing contextual information associated with the client device responsive to the request to generate the reply to the media content;
generating a plurality of reply suggestions based on the media attributes and the contextual information; and
causing display of the plurality of reply suggestions at the client device.

### Claim 11 (depends on 10)

The system of claim 10, wherein the receiving the request to generate the reply includes:
receiving an input that selects a graphical icon; and
generating the request to generate the reply based on the input.

### Claim 12 (depends on 10)

The system of claim 10, wherein the media content comprises a depiction of one or more objects, and wherein the generating the plurality of reply suggestions includes:
identifying the one or more objects based on the depiction of the one or more objects and the media attributes; and
generating the plurality of reply suggestions based on the one or more objects and the contextual information.

### Claim 13 (depends on 12)

The system of claim 12, wherein the identifying the one or more objects includes:
performing object recognition upon the media content; and
identifying the one or more objects based on the object recognition.

### Claim 14 (depends on 10)

The system of claim 10, wherein the generating the plurality of reply suggestions includes:
generating a prompt based on a predefined prompt template, the media attributes, and the contextual information;
providing the prompt to an artificial intelligence (AI) system; and
receiving, from the AI system, the plurality of reply suggestions.

### Claim 15 (depends on 14)

The system of claim 14, further comprising:
selecting the predefined prompt template from among a plurality of prompt templates based on one or more of the contextual information and the media attributes of the media content.

### Claim 16 (depends on 10)

The system of claim 10, wherein the causing display of the plurality of reply suggestions at the client device includes:
causing display of a first reply suggestion within a text input field at the client device;
receiving an input that comprises an input attribute;
responsive to the input, and based on the input attribute, causing display of a second reply suggestion within the text input field.

### Claim 17 (depends on 10)

The system of claim 10, wherein the media content comprises image data.

### Claim 18 (depends on 10)

The system of claim 10, wherein the contextual information includes one or more of:
location data;
temporal data; and
user profile data.

### Claim 19

A non-transitory machine-readable storage medium comprising instructions that, when executed by one or more processors of a machine, cause the machine to perform operations comprising:
causing display of a presentation of media content at a client device, the media content comprising media attributes;
receiving a request to generate a reply to the media content;
accessing contextual information associated with the client device responsive to the request to generate the reply to the media content;
generating a plurality of reply suggestions based on the media attributes and the contextual information; and
causing display of the plurality of reply suggestions at the client device.

### Claim 20 (depends on 19)

The non-transitory machine-readable storage medium of claim 19, wherein the receiving the request to generate the reply includes:
receiving an input that selects a graphical icon; and
generating the request to generate the reply based on the input.
