Meta bets on AI to replace human moderators as content policing shifts

Meta is gearing up for a revamp of the way it moderates content on its platforms. CEO Mark Zuckerberg supports plans to use more artificial intelligence systems, which could greatly reduce the need for human moderators, according to media reports.The drive centers on harnessing state-of-the-art AI agents and large language models (LLMs) to analyze posts, ads, and user activity on platforms like Facebook and Instagram. The systems are built to spot everything from spam and scams to harmful content, taking over a process that has traditionally depended on large teams of human reviewers.Meta has traditionally relied on a mix of automated tools and thousands of human moderators, many employed by third-party vendors, to review flagged content and enforce its community standards. These moderators made millions of decisions every day, including cases of hate speech, misinformation, violence, and fraud.Also read: Instagram tests new 'Your Algorithm' controls for feeds and Reels: What they mean for usersUnder the new approach, AI is expected to shoulder a much larger share of this workload. The company has already replaced a large chunk of human review processes with automated systems and plans to do so more as the technology improves, reports say.Internal tests detailed in the reports demonstrate that newer AI models can detect policy violations at scale and with greater speed and can more effectively identify patterns such as fraudulent accounts and coordinated scams over older methods.The move is part of a wider shakeup of Meta’s operations as the company pours money into artificial intelligence and tries to trim costs related to content moderation, which has long been one of the company’s biggest operational expenses. Meta spends billions of dollars a year on moderation, much of which goes to contractor services.Meta has said the shift is a multi-year rollout, not an immediate replacement of all human roles. Human moderators will still be involved in reviewing complex or context-driven cases, handling appeals, and overseeing the performance of AI systems.The development also signifies a significant shift in the way large social media companies view platform governance, with artificial intelligence now at the core of managing content across billions of posts and interactions daily.