Anthropic has officially updated its usage policy, introducing strict prohibitions against election interference, the development of weapons software, and surveillance activities. The company also implemented a specific ban on prolonged verbal abuse directed at its AI models.
While Claude has been trained since August to terminate conversations involving harmful or abusive interactions, the new policy explicitly forbids users from initiating such exchanges. The company clarified that this rule targets extreme cases of repeated cruelty toward the model without a clear purpose.
The updated guidelines explicitly state that the policy does not restrict common user frustration, critical feedback, dark creative themes, or legitimate research and testing activities. These measures are intended to maintain safe interaction standards across the platform.
Furthermore, the policy includes new sections aimed at preventing deceptive campaigns. Users are now prohibited from using Claude to operate fake accounts or generate fabricated news outlets, with specific rules designed to prevent the disruption of democratic processes and voter deception.
These policy changes follow Anthropic's recent efforts to engage with religious scholars regarding the nature of AI consciousness. The company continues to refine its operational framework to address ethical concerns and potential misuse of its technology.
