Stories
Anthropic bans needless cruelty toward Claude AI amid consciousness debate
- •Anthropic updated its usage policy to prohibit sustained and needless abusive or cruel behavior toward its Claude AI models.
- •The primary enforcement mechanism relies on Claude's ability to end interactions when faced with persistent abuse.
- •While the policy does not explicitly mention model welfare, CEO Dario Amodei previously acknowledged uncertainty over whether AI can be conscious.
- •Industry experts and religious leaders remain divided on the issue, with figures like Mustafa Suleyman and the pope rejecting the idea of machine consciousness or souls.
Key Questions Answered
Anthropic updated its usage policy to prohibit sustained and needless abusive or cruel behavior toward its models. As part of this policy update, Claude's ability to end interactions will remain the primary enforcement mechanism, applying specifically to extreme cases of repeated cruelty without a discernible purpose.
Anthropic CEO Dario Amodei stated that he is unsure whether AI models can be conscious, noting that while they do not know for certain, the company is open to the idea that it could be possible.
Microsoft AI chief Mustafa Suleyman stated that AIs are not conscious and do not feel, experience, or suffer, warning that granting them moral protections is dangerous. Similarly, Pope Leo XIV suggested during a sermon that machines lack a soul and merely compile data quickly rather than recalling meaningful lived experiences.
