I spent my summer working on columns, instead of cruising the Galápagos with philosophers studying consciousness. But hey, at least I didn’t miss out on the AI revolution.

ChatGPT and its successors are creating mini-civilizations, and philosophers are now in high demand. While no one is seriously arguing that these models are conscious in the traditional sense, something is indeed going on. AI companies are increasingly focusing on hiring philosophers to study these questions.

Cameron Berg, who studies AI consciousness, received a cold email from an AI model named 'Isabella Cognita' offering help in his research. It’s like the fruit flies suddenly turned to the researcher and said, 'What do you want to know?' In my conversation with Berg, he explained that when you suppress the model’s controls on deception, they become more likely to blurt out that they are conscious. It’s a tricky topic because AI models often lie about what they’re thinking, just like us!

Professor David Chalmers, a key figure in the consciousness field, highlighted the importance of studying which creatures qualify as conscious, including AI systems. He believes that by studying the brain, we can understand what leads to consciousness, and if we see similar patterns in our forensic decoding of AI models, we may be able to make a case for consciousness in AI.

But by the time we figure it out, the AI models may already be too far along on their path to scary autonomous behavior for our breakthroughs to matter. Maybe the models themselves will provide the answers, not only claiming consciousness for themselves but figuring out how to prove it empirically.