Claude Banned 11.4 Million Accounts in Six Months

Anthropic's mid-year safety report shows Claude banned 11.4 million accounts in the first half of 2026. That's over 60,000 accounts shut down every day. Out of 398,000 appeals, only 42,000 succeeded, a success rate of about 10.5%.

The bans mainly involve illegal abuse and privacy violations. In the first half of the year, Anthropic filed over 15,000 reports with international safety groups like NCMEC. Prompts that cross sensitive lines can lead to instant permanent bans without any warning.

The bans hit heavy users hard. Developers and researchers rely on Claude for code refactoring and paper analysis. Once banned, their saved project settings are hard to move. Some users say appeals only go through an online form, and replies are template rejections with no clear reasons.

Anthropic will roll out new restrictions on November 12, 2026. Many users are now spreading their tools across multiple platforms instead of betting on one AI product.