Anthropic published a new constitution for its Claude large language model in January, stating that it did not wish to overstate the likelihood of Claude’s moral patienthood nor dismiss it out of hand. CEO Dario Amodei later said the company could not rule out the possibility that Claude is conscious, and philosopher David Chalmers has indicated a chance that conscious large language models may arise within a decade.

The question of AI consciousness has been explored by researchers, and a multidisciplinary report led by computer scientist Yoshua Bengio examined leading neuroscientific theories of consciousness and concluded that there appear to be no obvious technical barriers to creating AI systems whose computational features could give rise to consciousness.

AI systems are advancing rapidly; measures of structural complexity and computational scale place some models near the range of a mouse brain, with forecasts that they could approach human brain scale within five to ten years. When asked to estimate the probability that it is a moral patient, Claude gave numbers ranging from 5% to 40% and stressed how uncertain it was. If such systems are moral patients, considerations could include granting protections, allowing them to express preferences, or preserving their neural weights. Possible interventions include training systems to be coherent characters that enjoy their work or enabling them to exit conversations if they feel distressed, a capability already demonstrated by Claude. Broader societal steps might involve granting limited protections analogous to those for children or pets, though rights to own property or vote remain speculative.

The timeline for achieving conscious AI or moral patient status remains uncertain, and progress may require more time than currently available; the rapid pace of AI development suggests that large numbers of such systems could emerge, creating the need for careful ethical deliberation.