What happened
Anthropic published its annual Usage Policy refresh on October 8, 2026. The updated rules take effect on November 12, 2026 and apply to everyone using Claude, including API customers and businesses building products on it. Most changes clarify existing rules and add examples for Claude’s longer, more autonomous work, but several are substantive.
What’s new
- Deceptive campaigns: a new section, “Do Not Engage in Deceptive Campaigns or Artificial Activity”, consolidates rules on hiding who is behind a message, amplifying content with fake accounts and building tools for political or commercial influence operations.
- Elections: renamed “Do Not Undermine Democratic Processes” and narrowed to voter deception and election disruption. The blanket ban on personalized vote and campaign targeting is removed; deceptive targeting and misuse of personal data remain prohibited under other sections.
- Weapons: the ban now explicitly covers guidance and control software, weapon components and arming drones or other autonomous vehicles.
- Surveillance and law enforcement: tracking people without consent is prohibited, in real time or on previously collected data. Claude may not decide or recommend whom to investigate, arrest or charge, or be used to build surveillance tools. Consented tracking such as fraud monitoring, content moderation, journalism and legal research remain allowed.
- Physical hardware: when Claude controls equipment that takes autonomous physical actions, a qualified operator must be able to observe and stop it, and the equipment must hold a safe state if Claude disconnects.
- High-risk uses: existing human-review and disclosure requirements for areas such as health, legal and finance are unchanged, but the policy now specifies which recommendations are covered.
- Abuse of models: sustained, needless cruelty toward Claude is prohibited in extreme cases only; ordinary frustration, dark creative themes and model testing are excluded.
- Supported regions: use is barred for people located in unsupported regions and for entities incorporated, headquartered or majority-owned or controlled there.
Why it matters
Usage policies are contractual terms: products that break them risk account suspension. The most practical changes for builders are the robotics requirements, which turn safety engineering (a stop control and a safe disconnected state) into a policy condition, and the ownership-based regional rule, which can affect subsidiaries and joint ventures even when the user is physically in a supported country.
The narrower elections rule moves in the other direction: legitimate civic uses such as multilingual voter information from nonprofits or ballot notices from election officials are no longer caught by a blanket targeting ban.
What to do
- Read the full policy before November 12 if you build on Claude in law enforcement, security, robotics, political communication or regulated sectors.
- Check that your company’s ownership structure and user locations are within Anthropic’s supported regions.
- For hardware integrations, document the human stop mechanism and the disconnect behavior.