Meta announced on Thursday that it is beginning to roll out more advanced AI systems to handle content enforcement as part of a plan to reduce its reliance on third-party vendors. These tasks include identifying and removing content related to terrorism, child exploitation, drugs, fraud, and scams.
The company stated it will deploy these new AI systems across its apps once they consistently outperform its current enforcement methods. While human reviewers will remain, Meta explained that the AI systems are better suited for repetitive reviews of graphic content or for areas where adversarial actors constantly change tactics, such as illicit drug sales or scams.
Meta believes these AI systems will detect more violations with greater accuracy, better prevent scams, respond faster to real-world events, and reduce over-enforcement. Early tests have shown promise, with the AI detecting twice as much violating adult sexual solicitation content as human review teams while reducing the error rate by over sixty percent.
The systems can also identify and prevent more impersonation accounts involving celebrities and other high-profile individuals. They help stop account takeovers by detecting signals like logins from new locations, password changes, or profile edits. Additionally, Meta says the AI can identify and mitigate approximately five thousand scam attempts per day, where scammers try to obtain user login details.
Meta emphasized that experts will continue to design, train, oversee, and evaluate the AI systems, making the most complex and high-impact decisions. People will still play a key role in high-risk decisions, such as appeals of account disablement or reports to law enforcement.
This shift occurs as Meta has been loosening its content moderation rules over the past year, following the start of a second term for President Donald Trump. Last year, the company ended its third-party fact-checking program in favor of a community notes model. It also lifted restrictions around topics in mainstream discourse and encouraged a personalized approach to political content.
The move also comes as Meta and other major tech companies face several lawsuits seeking to hold social media platforms accountable for harms to children and young users.
Separately, Meta announced the launch of a Meta AI support assistant, providing users with twenty-four seven support. This assistant is rolling out globally within the Facebook and Instagram apps for iOS and Android, and within the Help Center on the desktop versions of these platforms.