Anthropic Updates Usage Policy to Ban Model Abuse and Election Interference

Anthropic updated its usage policy to prohibit election interference, weapons software, and surveillance, and to bar deceptive campaigns such as fake accounts and fabricated news outlets. The policy also expressly forbids prolonged verbal abuse of Claude — a rule the company says applies only where users repeatedly act cruelly toward its models with no discernible purpose, and does not cover ordinary frustration, pushback, dark creative themes, or legitimate model testing and research.

The abuse clause follows an August update in which Anthropic trained Claude to end persistently abusive conversations on its own, part of a broader push by the company to define clearer boundaries for how its models can — and cannot — be used.