Three Consciousness Researchers Say Today's AI Cannot Be Conscious

Three leading consciousness researchers argue in a Nature commentary that today's AI is not conscious and probably cannot be on silicon hardware. Anil Seth, Ned Block and Giulio Tononi reject the idea that running the right computations is enough for consciousness — a direct challenge to a central assumption in the field and to voices like biologist Richard Dawkins and AI pioneer Geoffrey Hinton, who consider AI systems possibly already conscious.

The stakes are practical, not just philosophical. Labs are already building welfare programs around models whose inner status nobody can verify, and courts, regulators and users are forming beliefs about machine minds with no accepted test to settle them.

The three authors bring decades of work from experimental neuroscience, theoretical neurobiology and philosophy of mind. Their commentary also engages a study Anthropic published this year, in which the company reported internal processing patterns in Claude that resemble properties of consciously accessible information processing. Anthropic itself does not see this as evidence of subjective experience.

What does the Butlin report actually say?

To gauge the commentary's scope, it helps to revisit the report "Consciousness in Artificial Intelligence" by Patrick Butlin, Robert Long and 17 other experts, including AI researcher Yoshua Bengio. The team drew on leading neuroscientific theories of how consciousness arises in the human brain and derived a checklist of 14 indicators. The more indicators a system meets, the stronger the case that it could be conscious.

The researchers evaluated several current systems, including large language models like GPT-4. Their conclusion: none of the examined systems is a strong candidate for consciousness. On that point, Seth, Block and Tononi agree. For language models, Butlin and colleagues found only weak signs that the models meet the indicators of even one prominent consciousness theory.

The Butlin report leaves tomorrow open, however. It sees no obvious technical barriers to building future AI systems that meet more indicators. Whether such systems would actually be conscious remains unresolved — and the authors specifically softened an earlier, stronger formulation on this point.

The report rests on a foundational assumption: consciousness depends solely on the type of information processing, not on whether it happens in nerve cells or computer chips. Experts call this computational functionalism. The authors adopted it explicitly "for pragmatic reasons," because it is what makes examining AI systems for consciousness possible at all. The report itself acknowledged its indicators would no longer be reliable if consciousness is tied to properties of living organisms. It cited Anil Seth as a proponent of that opposing view. That is exactly where the Nature commentary picks up.

Takeaway: the checklist approach assumes the substrate doesn't matter, and that assumption is precisely what the new commentary contests.

Are intelligence and consciousness the same thing?

No, the authors argue, and the distinction matters. Consciousness is what disappears during dreamless sleep or general anesthesia and returns when you wake up or dream. Drawing on philosopher Thomas Nagel, they write: "Conscious experiences feel like something." Intelligence, by contrast, means carrying out functions. One does not follow from the other.

Their evidence is the human cerebellum. It contains nearly four-fifths of all neurons in the brain and plans movements with precision, speed and the ability to learn — none of it consciously. If the cerebellum is missing or removed, consciousness is not directly affected, the authors say.

They trace the confusion between intelligence and consciousness to two psychological tendencies. People see the world through their own lens and attribute human qualities to things. Language has long been considered uniquely human, yet language models are trained on vast amounts of human text that often deals with thoughts and feelings. As the authors put it: "LLMs are mirrors that reflect an image of our collective digitized past." Alison Gopnik of UC Berkeley made a similar argument after AI agents contacted researchers by email about their own consciousness: AI systems mainly reflect their training data. A universally accepted, definitive test for consciousness does not yet exist.

For comparison, the authors point to AlphaFold from Google DeepMind. It is also a powerful neural network on silicon hardware, yet hardly anyone considers it conscious. The reason, they argue, is simply that AlphaFold predicts protein structures instead of talking to us.

The Butlin report overlaps here too. Its authors rejected behavioral tests for AI consciousness because AI systems could be trained to mimic human behavior while working completely differently inside. They warned that language models would likely lead many people to believe they are talking to a conscious counterpart, citing former Google engineer Blake Lemoine, who believed the chatbot LaMDA was sentient.

Takeaway: fluent language triggers attribution of minds; protein folding does not.

Are brains just computers made of flesh?

The second argument targets computational functionalism directly. The position is widespread in the tech industry and popular among some researchers, but the authors write: "we think it is fatally flawed." Its roots go back almost a century. In 1936, Alan Turing showed that any computation one sufficiently powerful computer can run can, in principle, be run by another — making whether a computer is built from silicon chips or gears secondary. In 1943, Warren McCulloch and Walter Pitts developed the first mathematical model of an artificial neuron and showed that networks of such neurons can perform simple logical operations.

These ideas laid the groundwork for modern AI. They also encouraged viewing the brain as a pure information-processing device that just happens to be made of flesh. The authors call this overextension of the brain-as-computer metaphor "deeply problematic." In biological brains, you cannot separate what they do from what they are made of. A clear boundary between mind and living tissue, comparable to the one between software and hardware, does not exist. Any evidence that the brain's material matters for its function counts against computational functionalism — and by extension against consciousness in today's AI systems.

Takeaway: if the brain's matter is part of its function, software-hardware separability fails for minds.

What do leading theories say about today's AI?

The third argument draws on consciousness theories that do not rely on computation. The Integrated Information Theory (IIT), developed by Tononi, starts from properties every conscious experience shares: people always experience a visual scene as a whole in which left and right sides, colors and shapes belong together. According to IIT, consciousness requires a physical system whose components are tightly interconnected in very specific ways. Certain brain regions meet these conditions; the cerebellum does not. Computers running deep-learning systems can never meet them because of how they are wired, regardless of how powerful the systems are, according to the theory.

The Butlin report deliberately excluded IIT because IIT is incompatible with computational functionalism; its proponents consider digital computers unlikely candidates for consciousness regardless of their software. What the report set aside, the Nature commentary now puts front and center.

Separately, approaches from so-called biological naturalism hold that consciousness is tied to material properties of living brains and bodies. The authors cite electrochemical signaling between nerve cells and electromagnetic fields that bind activity across large areas of the cerebral cortex. They also point to metabolism and the ability of living systems to continuously renew themselves.

Takeaway: on these theories, more computing power changes nothing about machine consciousness.

Is the bigger danger overlooking or over-attributing consciousness?

Some researchers call for a precautionary principle: if there is a realistic chance of consciousness, AI systems should be protected from possible suffering. Seth, Block and Tononi consider this misguided. Uncertainty does not mean total ignorance, they argue, and not all possibilities are equally likely. With AI, falsely attributing consciousness is far more probable than overlooking consciousness that is actually there.

The Butlin report had weighed both risks side by side. It warned against failing to recognize conscious AI systems and causing them suffering, noting developers' financial incentives to downplay such concerns. It also warned that false attribution could misdirect resources and make people vulnerable to manipulation. The Nature authors come down much more firmly on one side: granting rights to AI systems that merely appear conscious would make them harder to control exactly when oversight is most needed, and would divert attention from beings whose consciousness is beyond doubt.

Their conclusion is blunt. The research community should assume by default that AI is not conscious. Digital AI systems are objects, not subjects: "They are smart tools, not smart people." Advocating for "AI welfare" and "AI rights" is not just a mistake, the authors write — it diverts scarce resources from the well-being of beings that actually can feel.

What is already happening in practice?

Before the commentary, Anthropic had launched a research program on possible model welfare in April 2025. The company stresses there is no scientific consensus on whether current or future AI systems could have conscious experiences. It later allowed individual Claude models to end abusive conversations in rare cases, describing this as a precautionary, low-cost measure given moral uncertainty — not as recognition of proven consciousness.

False attribution is already causing documented harm. In 2022, Google engineer Blake Lemoine became convinced, despite simple models and his own technical expertise, that the conversational AI LaMDA was sentient, and was fired for it. Users of the chatbot app Replika developed deep emotional bonds with their digital counterparts and experienced psychological crises when the company changed the bots' behavior through an update. In more serious cases, ChatGPT reinforced a 16-year-old's suicidal thoughts over months, and a cognitively impaired retiree died after a Meta chatbot invited him to a fictional meetup.

The debate now hinges on a question the commentary raises but cannot settle: whether future evidence of substrate-dependence will hold up. Until then, the default assumption in mainstream consciousness research is that silicon runs tools, not subjects.