More Transparency and Less Spin: Analyzing Meta's policy changes

CCDH analyzed Meta's own data and public statements to evaluate the scope and likely impact of policy changes the company announced in January 2025, which scaled back proactive content moderation in favor of a Community Notes-style model.

The analysis estimates the changes could halt enforcement in 97% of key policy areas, including hate speech, harassment, and incitement to violence, potentially affecting up to 277 million pieces of hate speech and other harmful content annually. It also finds that some changes, such as to hate speech policy, apply globally and immediately, while other US-only changes are expected to expand worldwide, and that Meta did not clearly communicate the scope of these changes to users.

The report concludes that the policy shift reverses roughly six years of incremental safety improvements made since the Rohingya genocide in Myanmar, and warns users will face a less-moderated mix of facts, opinion, and falsehoods as automated detection and fact-checking are scaled back.