Abstract
The proliferation of realistic, AI-generated content creates a significant risk of misinformation, as users often cannot distinguish it from human-created work. Existing detection tools are typically manual (i.e., copy content from one area to paste in a tool), disruptive for the user, and limited to a single content type. This disclosure describes a method integrated directly into a web browser for detection and visual indication of AI-generated content that can operate in real-time. When a user navigates to a web page, a content parsing module analyzes the Document Object Model (DOM). A multi-modal detection engine then scans content, such as text and images, for digital watermarks and other artifacts of AI synthesis. Identified content is then visually highlighted on the page. This provides users with a seamless and automated way to understand the origin of the content they are consuming, thereby increasing transparency and trust.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
Prakash, Abhay, "Real-Time Detection and Indication of AI-Generated Content in a Web Browser", Technical Disclosure Commons, (October 03, 2025)
https://www.tdcommons.org/dpubs_series/8673