Anthropic has recently updated its usage policy for its AI chatbot, Claude. The key changes include:
-
Abuse Prevention: The ban on the abuse of Claude is emphasized, along with new provisions addressing deceptive campaigns, elections, weapons development, surveillance, law enforcement, and hardware integration.
-
Deceptive Campaigns: A new section prohibits any form of deceptive activity, particularly in political or commercial contexts.
-
Elections: The rules now specifically state that users cannot use Claude to engage in deceptive practices that undermine democratic processes.
-
Weapons Development: The updated policy explicitly disallows using Claude to develop weapons or the technologies that support them.
-
Surveillance: The policy has tightened restrictions regarding surveillance and law enforcement, making it clear that Claude cannot be used for tracking individuals or any investigative recommendations.
-
Hardware Use: New requirements for Claude-connected hardware stipulate that operators must be able to intervene and ensure safety if needed.
The revised policy will take effect on November 12.
