Meta has removed approximately 10 million fake Facebook profiles in the first half of 2025 in a major crackdown targeting spam and impersonation. The accounts were found to be mimicking high-profile content creators and distributing unoriginal or AI-generated content that lacked authenticity.
The announcement was made by Meta on Monday as part of its renewed efforts to improve the relevance and trustworthiness of the Facebook feed. The company said the action is a step toward reducing spammy behavior, including mass content production using artificial intelligence tools.
In addition to the mass removal, Meta also took enforcement actions against 500,000 additional accounts identified as engaging in inauthentic behavior. These actions included demoting spammy comments and reducing the reach of duplicated or low-quality content — significantly affecting the ability of such accounts to monetize.
To support original creators, Meta has introduced stricter measures that reward and promote authentic and credited posts. It now uses AI-powered technology to detect duplicate videos and suppress their distribution in the feed.
This initiative aligns with Meta CEO Mark Zuckerberg’s announcement of increased investment in AI infrastructure, including plans to bring the company’s first AI supercluster online next year, with an investment commitment of “hundreds of billions of dollars.”
Meta’s actions come amid growing concern across platforms over “AI slop” — low-quality, repetitive content generated en masse by AI. YouTube also updated its policies earlier this month, stating that mass-produced or repetitive content would not qualify for monetization.
YouTube later clarified the policy isn’t anti-AI but targets unoriginal spam. “We welcome creators using AI tools to enhance their storytelling,” the company said in a blog post.
Meta and other tech giants are now under increased pressure to balance AI innovation with platform integrity, ensuring that original voices are amplified over automated noise.
