Anthropic changes usage policy to ban model abuse and election interference
techcrunch ·
Anthropic has updated its usage policy to explicitly prohibit model abuse and election interference. The new rules ban prolonged verbal abuse of the AI, a practice previously handled by training Claude to end such conversations but now strictly forbidden for users. This change targets extreme cases of cruelty without discernible purpose, excluding common frustration or creative testing. Additionally, the policy introduces strict guidelines against undermining democratic processes, including bans on deceptive campaigns and fake news generation. Other prohibitions cover the development of weapons software and surveillance tools. These updates follow recent collaborations where Anthropic engaged with religious scholars regarding the potential consciousness of its models. The company aims to balance safety measures with the need for open research and user interaction.
techcrunch