Seth, Block, Tononi: AI isn't conscious, may never be
TL;DR
- Anil Seth, Ned Block and Giulio Tononi argue in Nature that intelligence and consciousness are distinct, and today's AI has neither.
- The authors reject computational functionalism, the view that any hardware running the right program could be conscious, insisting substrate matters.
- They warn that wrongly attributing consciousness to AI could divert moral concern from beings with a genuine capacity to suffer.
Today's AI is not conscious, and may never become so, three leading consciousness researchers argue in a new Nature Comment. Anil Seth of the University of Sussex, Ned Block of New York University and Giulio Tononi of the University of Wisconsin–Madison set out three reasons, directly pushing back on more permissive views from figures including Geoffrey Hinton and Richard Dawkins.
The first is that intelligence and consciousness are different things. "Consciousness is about being and feeling; intelligence — very broadly — is all about doing, or performing functions," the authors write. They point to the human cerebellum, which contains most of the brain's neurons and handles fast, adaptable motor learning entirely without conscious awareness, as evidence that function and feeling can come apart. Large language models generate the appearance of inner experience, they argue, because they mirror "human-written text" that is itself concerned with thoughts, feelings and conscious experiences.
The second is substrate. The authors reject computational functionalism, the premise that the right computations yield consciousness on any hardware, writing that "brains are not — or at least not just — computers that are made of meat."
The third draws on specific scientific theories. Integrated information theory holds that consciousness requires tightly interconnected components, which the authors say the layered, feed-forward architecture of today's deep learning does not have. Biological naturalism goes further, pinning consciousness to material properties of living brains such as electrochemical synapses and metabolism. On either view, today's AI does not qualify.
The stakes are not academic. The authors warn that mistakenly attributing consciousness to AI could divert moral attention and resources from beings with a genuine capacity to suffer. The piece landed with momentum in the research circles we track — seven of the AI experts in our Who's Who directory shared it soon after publication.
Shared on Bluesky by 7 AI experts (top 5 by trust)
-
1/2 Three reasons why AI is not conscious - and perhaps never will be. New in @nature.com with Ned Block @neddo.bsky.social and Giulio Tononi www.nature.com/articles/d41...
View on Bluesky → -
Three reasons why AI is not conscious and perhaps never will be www.nature.com/articles/d41...
View on Bluesky → -
Great piece on why AI is not conscious, from people who really know what they’re talking about www.nature.com/articles/d41...
View on Bluesky →
Originally reported by nature.com
Read the original article →Original headline: Three reasons why AI is not conscious and perhaps never will be