Anthropic recently launched a groundbreaking research initiative called 'model welfare,' focused on the complex ethical questions around AI consciousness and moral consideration. The research areas include developing frameworks to assess consciousness, studying indicators of AI preferences and distress, and exploring interventions.

Anthropic appointed Kyle Fish, its first AI welfare researcher, who estimates a 15% chance that current models might already be conscious. The initiative builds upon increasing AI capabilities, as noted in a recent report co-authored by Fish, suggesting that AI consciousness could emerge soon.

Anthropic emphasizes the deep uncertainty surrounding these questions, noting that there is no scientific consensus yet on whether current or future AI systems could possess consciousness. The broader debate highlights a critical ethical frontier that society must begin addressing now.