Meta Expands AI Content Moderation, Cuts Reliance on Human Reviewers

Meta is beginning to deploy more advanced AI systems to handle content enforcement across its platforms, a shift that will reduce its reliance on third-party moderation vendors while expanding automation in how harmful content is identified and removed.

The company said the new systems will take on tasks such as detecting terrorism-related material, child exploitation, scams, and fraud, with broader rollout planned once performance consistently exceeds existing methods. Meta noted the technology is designed to handle areas where patterns change quickly or where reviews are repetitive, while human reviewers will remain responsible for more complex and high-risk decisions.

Early results suggest the systems are already improving performance. Meta said its AI can detect twice as much adult sexual solicitation content compared to review teams while cutting error rates by more than 60%. The tools are also being used to identify impersonation accounts and prevent account takeovers by flagging unusual activity, such as logins from new locations or profile changes. In addition, the company said it can now identify and block roughly 5,000 scam attempts per day.

The move reflects a broader effort to bring more of its enforcement infrastructure in-house. Rather than relying as heavily on external moderation partners, Meta is investing in systems trained on its own data and policies, aiming to improve accuracy and response times while reducing over-enforcement.

At the same time, Meta emphasized that human oversight will remain part of the process. “Experts will design, train, oversee, and evaluate our AI systems, measuring performance and making the most complex, high-impact decisions,” the company said, adding that people will continue to handle sensitive cases such as appeals and law enforcement reports.

The shift comes amid wider changes to Meta’s content policies over the past year, including the end of its third-party fact-checking program and a move toward a Community Notes-style model. It also arrives as the company and other social platforms face ongoing legal scrutiny related to user safety, particularly involving younger audiences.

Alongside the enforcement updates, Meta also introduced a new AI-powered support assistant. The tool is rolling out globally across Facebook and Instagram on mobile and desktop, providing users with 24/7 help through the apps’ Help Center.

Together, the updates signal a restructuring of how Meta manages moderation, combining increased automation with continued human review as the company scales enforcement across its platforms.

This analysis is based on reporting from TechCrunch.

Image courtesy of Meta.

This article was generated with AI assistance and reviewed for accuracy and quality.