Meta Pushes To Label AI Images on Social Media Sites

Meta will begin detecting and labeling images generated by other companies’ artificial intelligence services using a set of invisible markers built into the files, according to Nick Clegg, president of global affairs, in a blog post on Feb. 6, 2024.

Meta has said the move comes during a year when a “number of important elections are taking place around the world.” Therefore, the use of AI images is expected to become a more “adversarial space in the years ahead.”

In the blog post, Clegg wrote, “We’ve been working with industry partners to align on common technical standards that signal when a piece of content has been created using AI.” This includes both audio and video. “Being able to detect these signals will make it possible for us to label AI-generated images that users post to Facebook, Instagram and Threads.” Photorealistic images created using Meta AI will be labeled “Imagined with AI.”

Clegg also stated, “As the difference between human and synthetic content gets blurred, people want to know where the boundary lies. People are often coming across AI-generated content for the first time and our users have told us they appreciate transparency around this new technology.”

He continued by explaining how important it is to tell users when images are photorealistic via their Meta AI feature or created using other companies’ tools.

Clegg added that he understands it is only possible to identify some AI-generated content. He also knows the invisible markers noted on questionable images can be erased.

However, Meta is confident that it can find ways to make removing or altering invisible watermarks more difficult. Clegg concluded that it is up to social media users to determine if AI has created the content, like checking whether the account sharing the content is trustworthy or looking for details that might look or sound unnatural. 

Meta’s announcement is just the beginning of an emerging system of standards technology companies are developing to mitigate the potential harms of generative AI technologies.