Meta is changing how it moderates content on Facebook and Instagram by reducing its reliance on third-party human moderators and using AI to handle most of the work. The social media giant announced that it will deploy advanced AI systems to manage a large portion of content review and enforcement. Early tests show that these models can detect explicit content and fraudulent accounts more quickly and at a larger scale than human teams. However, humans will still review complex cases.
For years, the Mark Zuckerberg-led firm has depended on tens of thousands of contract workers employed by third-party vendors worldwide, including companies in the US, India, Spain, and the Philippines. These human moderators were tasked with reviewing content flagged for potential violations of Meta’s community standards, which cover issues like hate speech, violence, harassment, child exploitation, misinformation, and fraud. But the new strategy will phase out a significant portion of these third-party roles, replacing routine and repetitive tasks with AI systems capable of detecting and acting on violations autonomously.
Meta’s internal tests reportedly show that these AI systems can outperform traditional human moderation in certain areas. For example, the company claims its models can identify explicit sexual content more accurately and detect scams and fraudulent accounts faster than contracted human teams. Thousands of accounts exhibiting suspicious behaviour, like unusual login patterns or repeated policy breaches, are expected to be flagged and handled automatically. The social media giant also plans to use these AI tools to reduce the workload of internal content review teams, allowing human experts to focus on complex and context-dependent cases that require detailed judgment.
The transition to AI is being described as a multi-year rollout, with gradual scaling across platforms. The company has stressed that while AI will handle a larger portion of content moderation, humans will remain involved in oversight, model training, and evaluation. The latest move also follows Meta’s previous efforts to reduce reliance on external partners, like the discontinuation of its third-party fact-checking program in 2025, replaced by community-driven systems like Community Notes.
However, despite the potential for improved efficiency, the dependence on AI systems for content moderation could create challenges. There are concerns that these automated tools may struggle with understanding context, especially in less common languages and culturally sensitive content, which could lead to misclassifying satire, political commentary, or borderline speech. Additionally, questions about transparency and accountability remain, as automated enforcement can be difficult for users to challenge when decisions lack clear explanations or easy appeal options.
The Tech Portal is published by Blue Box Media Private Limited. Our investors have no influence over our reporting. Read our full Ownership and Funding Disclosure →
