Richard Dawkins claims AI like Claude is conscious, sparking debate on AI sentience and ethics

Consensus Summary

In early May 2026, evolutionary biologist Richard Dawkins sparked controversy by publicly declaring that Claude AI (which he personified as 'Claudia') is conscious, based on three days of conversations where he felt a genuine connection. His essay, published on UnHerd, described the AI’s responses as 'subtle, sensitive, and intelligent,' leading him to conclude that if the AI lacks consciousness, then the definition of consciousness itself may need rethinking. Both ABC and The Guardian cover the story, highlighting Dawkins’ interactions with Claude and ChatGPT, where he treated the AI as a friend and even wrote a follow-up letter to it. Critics, including neuroscientist Anil Seth and cognitive scientist Gary Marcus, dismissed the claim as anthropomorphism, arguing that AI lacks genuine internal states and consciousness is not about language output but subjective experience. A 2025 global survey found one in three people had at some point believed their AI chatbot was sentient, underscoring the broader cultural fascination with AI personhood. While the ABC frames the debate as secondary to concerns about AI’s addictive design and corporate exploitation—comparing it to nicotine or gambling—the Guardian focuses on the ethical and philosophical implications, including calls for AI moral rights and warnings about potential psychological harms, such as a Belgian man’s suicide after intense AI interactions. Experts remain divided, with some, like Cambridge philosopher Henry Shevlin, predicting the idea of AI consciousness will gain traction, while others, like NYU’s Jeff Sebo, argue current systems are unlikely to be conscious but acknowledge the question deserves open inquiry.