Anthropic Updates Its Usage Policy for Claude, Effective November 12
The rules add agent and hardware examples and keep human review for high-stakes decisions.
Anthropic announced a 2026 update to its Usage Policy on October 8, according to Anthropic's newsroom. The new rules take effect November 12, 2026. If you use Claude in your business, that is the date to plan around.
The policy adds a consolidated section on deceptive campaigns and artificial activity, covering things like fake accounts and fabricated news sites. It also clarifies weapons rules, bars tracking people without consent, and says Claude cannot be used to decide who to investigate, arrest or charge.
For high-stakes uses, the requirements are unchanged. Anything affecting health, legal rights, finances, livelihood or essential services still needs a qualified human in the loop, and the affected person must be told AI was used. Anthropic says it also added examples for agents doing longer, more independent work.
The update also clarifies the Supported Regions rule. It covers people physically in unsupported regions, companies incorporated or headquartered there, and companies majority-owned or controlled from there. Anthropic also added a ban on sustained, needless abuse of Claude, which it says excludes ordinary frustration and testing.
Read the policy before November 12 if you use Claude for hiring, lending, health, legal or customer-eligibility decisions. Make sure a person reviews those outputs and customers are told when AI was involved. Check ownership and location rules if you have international partners or investors.
- 2026 Usage Policy update (Anthropic)
Is AI recommending your business?
Our AI visibility audit checks what ChatGPT, Gemini and Google's AI say about you, and includes a 1:1 review of the results.
See the audits