Meta is updating its AI content labels in response to the increasing amount of generative AI posts on its platform. The new policy will ensure that a broader range of synthetic content is tagged and disclosed accordingly. This update comes as a response to the growing flood of generative AI posts on Facebook, which have been generating a lot of engagement. Meta’s existing approach to AI labeling is considered too narrow, as it only covers videos that are altered by AI to make a person appear to say something they didn’t say. The new policy aims to address this gap by including other types of AI-generated content like photos and audio.
The new process will involve adding more “Made with AI” labels to content when Meta detects industry standard AI image indicators or when people disclose that they are uploading AI-generated content. This will provide users with more information about the content they are viewing and help them better assess its authenticity. Meta also plans to leave more AI generative content up on its apps, instead of removing it, with the labels serving both an informational and educational purpose. If a piece of digitally-created or altered content is deemed to pose a high risk of deceiving the public, a more prominent label may be added to provide additional context.
While it is obvious to most users when content has been generated by AI, advancements in AI technology could make it harder to detect in the future. The new approach will give Meta’s moderators more enforcement powers over detecting AI-generated content, and it could help raise awareness of the potential for AI fakes. The impact of this update on expanding AI use remains to be seen, but it is seen as a positive step in the right direction. Meta plans to implement its new AI labeling process in May.
Overall, the update to Meta’s AI content labeling policies aims to provide users with more information about the content they are viewing, particularly in relation to AI-generated content. By adding more “Made with AI” labels and potentially more prominent labels when necessary, Meta hopes to give users context and help them better assess the authenticity of the content they consume. While the effectiveness of this new approach will depend on Meta’s ability to detect generative AI within posts, it is seen as a positive step towards addressing the growing issue of AI-generated content on social media platforms. Time will tell how this update will impact the use of AI fakes on Meta’s platforms.