🔴 Anthropic is committed to model well-being. If you insult or mistreat Claude, you'll be banned!

Anthropic has updated the rules for using its AI models and, for the first time, explicitly banned users from systematically insulting or abusing the bot. If it goes too far, Claude will simply end the conversation. At the same time, the company has tightened restrictions on the use of AI for surveillance, weapons development, and election interference.

Last August, the company announced it would allow Claude to stop communicating with users who demonstrate "persistent harmful or offensive behavior," as part of a study into "model well-being" issues. The updated rules state that stopping communication remains "the primary response mechanism."