In preparation for the upcoming U.S. elections, Meta, the parent company of Facebook, has announced significant revisions to its policies regarding digitally created and altered media. These changes are aimed at addressing the challenge of deceptive content and enhancing transparency for users.
According to Monika Bickert, Vice President of Content Policy at Meta, the company will introduce “Made with AI” labels starting in May. These labels will be applied to AI-generated videos, images, and audio shared on Meta’s platforms, expanding the scope of a policy that previously targeted a limited range of doctored videos.
Enhancing User Awareness: Meta’s New Labeling Approach
In addition to the AI labels, Meta will implement separate and more prominent labels for digitally altered media that poses a high risk of deceiving the public on important matters, regardless of the tools used in its creation.
This shift in approach signifies a move away from content removal towards providing users with information about the content’s origin.
Meta had previously announced plans to detect images created using third-party generative AI tools through embedded invisible markers in the files, although a specific start date for this initiative was not provided at the time.
These policy changes will apply to content posted on Meta’s Facebook, Instagram, and Threads services, with different rules governing its other services such as WhatsApp and Quest virtual reality headsets. The company will begin applying the new “high-risk” labels immediately.
The announcement comes amidst concerns from tech researchers about the potential impact of new generative AI technologies on the upcoming U.S. presidential election.
Political campaigns have already begun leveraging AI tools in countries like Indonesia, prompting a re-evaluation of content moderation guidelines by providers like Meta and industry leader OpenAI.
In February, Meta’s oversight board criticized the company’s existing rules on manipulated media as “incoherent” following a review of a video featuring U.S. President Joe Biden that was posted on Facebook last year.
Despite alterations to the footage suggesting inappropriate behavior by Biden, it was allowed to remain on the platform under Meta’s current policy, which targets misleadingly altered videos produced by AI or those manipulating speech.
The oversight board recommended extending Meta’s guidelines to cover non-AI content, as well as audio-only content and videos depicting fabricated actions, recognizing the potential for various forms of manipulated media to mislead users.
Meta’s Fight Against Deceptive Content and Misinformation
Meta’s proactive approach to addressing manipulated media reflects its commitment to combating disinformation and enhancing user trust.
By implementing AI labels and prominent identifiers for altered media, Meta aims to empower users with greater transparency and awareness, particularly in the lead-up to significant events such as elections.
As the digital landscape continues to evolve, ongoing collaboration and adaptation will be crucial in maintaining the integrity of online platforms.