Meta Platforms plans to transition over 90% of human content review requests to large language models by the end of 2026. The company already handles approximately 50% of these moderation tasks using AI.
This initiative aims to reduce operating costs and automate core processes across Facebook and Instagram. These platforms currently serve nearly 4 billion monthly active users.
Meta expects AI-driven moderation to improve the speed and accuracy of identifying harmful material. The company also reportedly paused an internal employee tracking program used to gather data for training its AI models.
Management views this shift toward AI efficiency as a key driver for future growth and profitability. This strategic pivot reflects a broader automation trend within the communication services sector.