Meta Platforms, the parent company of Facebook, is collaborating with industry peers to establish standardized practices and tools for identifying content manipulated by artificial intelligence (AI). In an effort to address this issue, Meta announced its plans to develop tools capable of detecting invisible markers associated with AI-generated content. This development aims to enable proper labeling when such images are posted on Facebook, Instagram, and Threads.
Meta's ongoing efforts in building this capability intend to facilitate the application of labels in the near future. Invisible markers have already been incorporated by some companies in their image generators. However, Meta acknowledges that there is still progress to be made regarding audio and video content generated at the same scale. As the industry works towards this goal, Meta plans to introduce a feature that allows users to disclose when they share AI-generated video or audio. This disclosure would enable Meta to appropriately label these types of content as well.
Although not all AI-generated content can currently be identified, Meta acknowledges that there are methods to remove or modify invisible markers associated with AI. Consequently, Meta is focused on enhancing the robustness of these markers, making them harder to alter or eliminate. Additionally, Meta is actively developing capabilities for automatically detecting AI-generated content independent of invisible markers.
Meta recognizes that the proliferation of AI-generated content is still in its early stages. However, as it becomes more prevalent in the years to come, Meta anticipates societal debates regarding the identification and regulation of both synthetic and non-synthetic content.
In an attempt to tackle the challenges posed by AI-generated content, Meta Platforms is taking a proactive approach and collaborating with industry peers. By developing tools capable of detecting invisible markers associated with such content, Meta aims to ensure appropriate labeling on its platforms. As the industry continues to advance in this domain, Meta plans to enhance detection capabilities and facilitate user disclosure for AI-generated video and audio.