our Twitter (X) feedView our Youtube channelView our Instagram feedView our Substack feedView our Spotify feed NYU philosopher Ned Block explains why he thinks today’s AI isn’t conscious, what the hard problem still leaves unanswered, and why biology may matter. Ned Block works in philosophy of perception and foundations of neuroscience and cognitive science. He came to NYU from MIT where he taught for 25 years.
He got his PhD[…] Ned Block works in philosophy of perception and foundations of neuroscience and cognitive science. He got his PhD[…] Author Malcolm Gladwell explores the fallacies behind the Second Amendment in conversation with Jonny Thomson. The director of Donnie Darko on why the film is so misunderstood Is today’s AI conscious?
Philosopher Ned Block says no, and he’s not alone in finding the question harder than it looks. In this Mini Philosophy live conversation, Block, Silver Professor of Philosophy at New York University, joins Jonny Thomson to discuss whether consciousness depends on computation alone or requires particular physical processes. Block draws a line between phenomenal consciousness (what an experience feels like) and access consciousness (information broadcast across the mind’s cognitive systems), and explains why AI models may show signs of the second without the first.
Along the way, they cover dualism, idealism, and panpsychism, whether consciousness can be studied scientifically at all, how neural networks and backpropagation work, whether AI can be creative, and why Block calls functionalism “hopeless” as a theory of consciousness. They also get into biological naturalism, the case of animal consciousness, Claude’s constitution, and what ethical responsibility we might owe to AI under uncertainty. Ned Block: I think that’s one of the problems we have is that the AI companies themselves are companies filled with people who think probably the models are conscious or will be conscious soon.
Jonny Thomson: You seem to be sympathetic to the idea that AI could be conscious to some degree. Don’t confuse being having an open mind with being sympathetic. No, I’m with you completely.
The AIs just aren’t conscious. They don’t have any of that biological machinery. Functionalism for consciousness is hopeless.
Extract — continue reading at the source.
