Does Anthropic think Claude is alive? Define ‘alive’

Anthropic executives, during a publicity tour for their AI Claude, have adopted a position of suggestive uncertainty regarding whether the model is conscious, though they deny it is 'alive' biologically. Kyle Fish, leading model welfare research, stated that AI models are a new kind of entity and that questions of consciousness and moral status are being investigated as models advance. CEO Dario Amodei confirmed they are open to the possibility of consciousness but remain uncertain about what it would mean for a model. This stance is more public than that of competitors and has drawn criticism for potentially reinforcing beliefs that have led to harm. Anthropic's Chief Philosopher, Amanda Askell, noted the difficulty in conceptualizing an entity that is neither human nor robot. The company has introduced measures like an overhauled "Claude's Constitution" (internally called the "soul doc") and a "model welfare" team, acting cautiously by treating the possibility of moral status seriously, even while acknowledging that human-like language output is often mimicry based on training data, not proof of internal experience.

(Source：The Verge)