Anthropic has updated its usage policy for Claude for the first time in over a year to explicitly ban users from systematically abusing the AI assistant. Under the updated terms, sustained and needless abusive or cruel behavior toward Claude is prohibited, with violations potentially leading to warnings, throttling, restrictions, suspensions, or account termination. The company clarifies that the rule is meant for extreme cases and does not apply to normal user frustration, pushback, dark creative themes, or model testing and research.
This policy update aligns with Anthropic's broader approach to treating Claude as more than a simple tool, reflecting ongoing research into AI model welfare and potential moral patience. Alongside the anti-abuse rule, the updated usage policy also expands bans on propaganda campaigns using fake accounts, broadens weapons restrictions to cover software and drone weaponization, and explicitly prohibits non-consensual surveillance while carving out potential exceptions for government contracts.
Why it matters: This policy shift highlights how leading AI labs are increasingly defining behavioral boundaries not just for system safety, but around how humans interact with models perceived to have functional emotional states.