The role of the Facebook fact-checker page in detecting AI-generated fake content

Abstract

This work investigates the role of the Facebook page "The Checker" in detecting fake content produced using artificial intelligence techniques. This type of content has become increasingly prevalent in recent years due to the ease of sharing artificially generated or manipulated images and videos. The study focused on monitoring the content published by the page between January 1, 2023, and December 29, 2024, to identify the nature of the misleading news that was verified and the methods the Checker team used to detect it.

The researcher employed a descriptive-analytical approach, analyzing a comprehensive sample of 20 pieces of fake content. These were categorized according to their type, format, reach, and degree of manipulation. The results showed that political content was the most prevalent, followed by social and religious content. The text-to-image format was the most widely shared among users. Furthermore, the majority of the verified news items were international in nature, and most fell under the category of total fabrication, while deepfakes constituted the largest proportion of manipulation cases.

The results showed that the investigation team relied primarily on digital verification tools and direct communication with sources to refute claims, and that social media was the primary source of the spread of fake content. The study also demonstrated that the platform effectively contributes to reducing the spread of rumors by providing comparative evidence and clarifying the true contexts of posts.