Meta is expanding its use of artificial intelligence to handle content enforcement across its apps, as the company plans to reduce reliance on third-party moderation vendors while improving how it detects harmful content like scams, fraud, and illegal activity.
The company says these advanced AI systems will take over repetitive and high-volume moderation tasks, allowing faster detection of violations such as impersonation accounts, scam attempts, and harmful content, while human reviewers continue to handle complex decisions and appeals.
Meta confirmed the changes in a blog post, where it outlined how AI will support enforcement at scale. “While we’ll still have people who review content, these systems will be able to take on work that’s better-suited to technology,” the company said.
Smarter AI for Content Enforcement
Meta says early testing shows strong results, with AI systems detecting more violations while reducing mistakes.
- Detects twice as much violating adult solicitation content
- Cuts error rates by over 60%
- Identifies around 5,000 scam attempts daily
- Reduces impersonation reports involving celebrities by over 80%
- Helps prevent account takeovers by spotting unusual login patterns
These systems also support more languages and can quickly adapt to changing tactics, slang, and region-specific behavior used by bad actors.
Meta plans to scale these AI tools over the next few years and reduce its dependence on external moderation vendors. The company says human experts will still design, monitor, and evaluate the systems, especially for high-risk decisions like account appeals and legal reports.
“Experts will design, train, oversee, and evaluate our AI systems,” Meta said.
New AI Support Assistant Rolls Out Globally
Alongside enforcement changes, Meta has launched a new AI support assistant on Facebook and Instagram, offering 24/7 help for account issues.
- Reset passwords and update profile settings
- Report scams and impersonation accounts
- Manage privacy controls
- Track content removal and appeal decisions
The assistant responds in seconds and is now available globally across mobile apps and desktop Help Center, with more features planned over time.
Meta says these updates aim to improve safety, reduce harmful content, and make support faster and easier for users.
