Anthropic updates Claude policy, may ‘ban’ users who are abusive or cruel to its AI

Artificial intelligence company Anthropic has updated its acceptable usage policy for the Claude AI model to prohibit users from directing sustained, needless abuse or cruelty toward the system. Announced on Thursday in San Francisco, the refined guidelines are designed to combat various forms of mistreatment directed at the AI while also addressing emerging threats such as deception and election interference.

The policy update arrives amid an ongoing philosophical debate regarding whether artificial intelligence systems are capable of consciousness. By implementing these usage restrictions, Anthropic aims to establish behavioral standards for human interactions with its AI models.