Meta announced today that it will begin labeling content uploaded to Facebook, Instagram, and Threads if it was created using AI. The new policy aims to make it easier for users to identify images generated by AI tools without the need for human review.
In the coming months, Meta will watermark photos generated by its own generator and those of other major AI tools such as Midjourney, Dall-E, and Bing Image Creator. The watermarks will read "Imagined with AI" to clearly indicate that the image is not an authentic photo but rather AI-generated media.
It’s important that we help people know when photorealistic content they’re seeing has been created using AI. We do that by applying “Imagined with AI” labels to photorealistic images created using our Meta AI feature, but we want to be able to do this with content created with other companies’ tools too.
In addition, Meta will soon require users to disclose if they are sharing realistic video or audio created solely with AI. Failure to provide this information upfront could result in penalties such as warnings or even removal of the post.
Nick Clegg, president of global affairs at Meta, said the company needs to be "pretty vigilant" about AI-manipulated media being shared around election time that is intended to deceive the public.
If we determine that digitally created or altered image, video or audio content creates a particularly high risk of materially deceiving the public on a matter of importance, we may add a more prominent label if appropriate, so people have more information and context.
Full detection of synthesized video and audio is still a challenge that Meta can"t solve on its own. The company says it is working closely with partners to help detect and label synthesized media in real-time, but there may still be challenges with content that evades detection.
The move is part of an ongoing effort to develop industry standards for the transparent labeling of AI content. In a similar move, YouTube announced plans in 2023 to help its billions of users find out if a video they"re watching was created with generative AI tools.