Meta Platforms operates Facebook, Instagram, WhatsApp, and Threads. It is one of the world's largest AI research and deployment organizations, with extensive AI use in content moderation, advertising, and feed ranking.
Meta's AI-powered feed ranking systems determine the order of content in Facebook and Instagram feeds, training on engagement signals that have been linked in leaked internal research to increasing anxiety and polarization in some users.
Meta uses deep learning ranking models that optimize for engagement metrics. Internal research leaked by whistleblower Frances Haugen showed Meta's own researchers found the ranking system can amplify divisive content.
Meta uses AI to automatically detect and remove violating content across Facebook and Instagram, handling billions of posts per day. The AI acts before human reviewers in the majority of proactive removals.
Meta's content moderation AI handles categories including hate speech, violence, nudity, misinformation, and spam. According to Meta's transparency reports, AI is now responsible for detecting over 90% of policy-violating content before users report it.
Have evidence about Meta's AI practices? Submit it.
Report a Sighting