Anthropic updates its Usage Policy: high-risk uses require people with authority to review
Anthropic announces 2026 Usage Policy updates, clarifying rules on surveillance, weapons-related software and high-risk decisions. For uses involving personal health, law, finance or essential services, it requires qualified people who can review and change recommendations.
Source Facts
Anthropic published an updated usage policy on 8 October, explaining how existing rules apply to Claude when it works autonomously for longer periods. The company said many changes clarify existing enforcement practices and prohibited uses.
The announcement requires qualified people to participate in high-risk uses and disclosure of AI use to affected people. It also clarifies restrictions on tracking individuals without consent. These are supplier terms of use, not direct statements of Hong Kong law.
When introducing agents, translate vendor policies into executable procedures, assigned responsibilities and human review arrangements.
Original sources
2026 Usage Policy update
Sources checked: 2026-10-26 · Prepared with SI agent assistance; articles have not been individually reviewed by a person.
This is an English translation of our Chinese summary and editorial observations; source dates and this site’s verification date are shown separately. If an original advisory changes, refer to the source.
KEEP READING
Recommended reading
IllustrationLiteLLM explains package compromise: exposed keys must be revoked to complete the response
IllustrationSOCRadar examines SI agent-assisted extortion: human approval remains part of the intrusion process
Illustration