Anthropic Updates Usage Policy to Prohibit Model Abuse and Election Interference
Anthropic has revised its usage policy to explicitly ban prolonged verbal abuse of its AI model, Claude, alongside new prohibitions on election interference, weapons software, and surveillance. The updated policy clarifies that while common user frustrations or creative themes are permitted, persistent and cruel interactions with no discernible purpose are now forbidden. This change follows an August update where Claude was trained to terminate conversations involving persistently harmful or abusive user interactions. Additionally, the policy includes sections titled 'Do Not Undermine Democratic Processes,' which specifically forbid using Claude for deceptive campaigns, such as running fake accounts, fabricating news, deceiving voters, or disrupting elections. These updates reflect Anthropic's ongoing efforts to ensure its AI systems are used safely and ethically.
