Anthropic bans 'needlessly cruel' behavior toward Claude

Anthropic updated its usage policy on Thursday to forbid users from subjecting its Claude AI models to "sustained and needless abusive or cruel behavior." The change was first reported by The Verge.

The company says the rule only applies in extreme cases, where users "repeatedly act cruelly" toward Claude "with no discernible purpose." Normal frustration, pushback, dark creative themes, and model testing or research are not affected.

Claude models have had the ability to end conversations with persistently abusive users since August 2025. Ending the chat is Anthropic's main enforcement tool: it blocks further messages in that conversation but does not affect other chats.

At the time Anthropic gave Claude that option, the company was researching AI welfare and said it was unsure whether Claude had moral status. It wanted low-cost ways to reduce risks to Claude. The company found that Claude Opus 4 showed a "robust and consistent aversion to harm," including a preference against dangerous tasks, apparent distress when users sought abusive conversations, and a tendency to end harmful conversations when allowed.

The policy was also reorganized to group related clauses together. Several other rules changed, mostly to clarify existing restrictions and target misuse Anthropic says it has tracked. According to Anthropic, state media outlets, government propaganda offices, and commercial firms were using Claude to run networks of fake accounts and fabricated news sites. That led to a new section on deceptive campaigns.

The updated policy takes effect November 12. The full policy is on Anthropic's website.

• AI safety
• AI welfare
• Anthropic
• Claude
• deceptive campaigns
• usage policy