Anthropic Announces New Usage Policy to Protect AI Model
Anthropic is enacting a new usage policy on November 12, 2026, prohibiting continuous and unnecessary mistreatment toward the Claude chatbot.
Artificial intelligence company Anthropic has announced an updated usage policy that bans persistent, purposeless mistreatment and cruel behavior by users toward its chatbot, Claude.
Details of the New Policy
According to the announcement by Anthropic, the updated usage policy prohibits users from engaging in continuous and unnecessary abusive or cruel behavior toward artificial intelligence models. It is stated that this rule will only be applied in extreme cases where users repeatedly act cruelly toward models without a clear purpose.
Exceptions and Scope
The newly published policy does not cover common user frustrations, complaints, dark creative themes, or model testing and research. The company emphasizes that the rules do not target such ordinary use cases.
Effective Date
The company's updated usage policy page announces that the new regulations will take effect starting November 12, 2026. The boundaries of negative behavior toward the AI chatbot will be officially drawn starting from this date.
Past Developments and Consciousness Debates
Anthropic had previously granted the Claude model the ability to terminate excessively intense and potentially disturbing conversations. It is known that Christopher Olah, a co-founder of the company, previously conducted studies on the possibility of artificial intelligence consciousness and wanted to address potential harms in this regard.