Which operation can computers (including quantum computers) not perform, that human neurons can? If there is no such operation, then a human-brain-equivalent computer can be built.
>it is argued that the human mind cannot be computed on a Turing Machine... because the latter can't see the truth value of its Gödel sentence, while human minds can
And the debunk is that both Penrose and an LLM can say they see the truth value and we have no strong reason to think one is correct and the other is wrong. Either of both could be confused. Hence the argument doesn't prove anything.
Having read about Penrose's positions before, this is indeed what is he proposing in a roundabout way: that there is an origin to "consciousness" that is for all intents and purposes metaphysical. In the past he pushed the belief that micro-tubules in the brain (which are a structural component of cells) act like antennas that receive cosmic consciousness from the surrounding field.
In my opinion this is also Penrose's greatest sin: using his status as a scientist to promote spiritual opinions that are indistinguishable from quantum woo disguised as scientific fact.
Like I said I have no business talking about philosophy or spiritualism. However, since you asked: that's not at all what I meant. In fact, it's the opposite way around. I'm of the opinion just because we don't know something, this shouldn't give people a license to invent things from whole cloth and assert them as facts (which is exactly what Penrose does).
We're still waiting on proof of anything supernatural, and explaining things with materialism has served us super well. It's not unreasonable to assume it's going to continue to be a good tool for understanding the world.
I believe Penrose's core argument fits the description of a rhetorical device called argument from incredulity. He is incredulous how "consciousness" could ever arise from mere molecules interacting with each other. To me, everything he built up on top of this is tantamount to intellectual dishonesty, but I acknowledge that this is born out of a certain bias on my end.
1. Computers cannot self-rewire like neurons
2. No computer operates with the brain’s energy efficiency
3. Human learning is continuous and unsupervised, which is not possible for any computer
Computers don't need to "rewire" themselves, since neurons aren't implemented directly in hardware. When you do RLHF, the parameters inside the model are "rewired" in the sense that is relevant for the purpose of this discussion.
> No computer operates with the brain’s energy efficiency
No existing human-made computer operates with the brain's energy efficiency, true. But the premise isn't about specific computers, but computers in general. There's no reason to believe that a computer operating with the same efficiency is impossible. The efficiency of the human brain is still well within the limit imposed by thermodynamics, and everything above that limit is, in principle, possible.
> Human learning is continuous and unsupervised, which is not possible for any computer
This is just plainly not true. Continuous learning with existing LLMs is trivial (just too expensive to actually bother). Unsupervised learning is literally how LLMs are trained initially.
Neural synapses can physically grow, shrink, change receptor densities, and form new pathways dynamically and autonomously at multiple timescales.
The brain adapts neuron-by-neuron based on local conditions (e.g. a single neuron strengthens its connection based on local neurotransmitter activity), RLHF adjusts millions of parameters in bulk, requiring external training loops, gradient descent, and centralized loss functions — nothing like self-rewiring at an individual unit level.
>There's no reason to believe that a computer operating with the same efficiency is impossible.
Theoretically possible, yes, but no current computational paradigm operates with anywhere near the efficiency of biological neurons; there is no reason to believe it will change in any foreseable future. If you think we know even 1% of the laws which hold it all together, well, you are very human and also a big optimist.
>This is just plainly not true. Continuous learning with existing LLMs is trivial
The claim isn’t about feasibility but about how continuous learning in AI is fundamentally different from human learning. AI models cannot learn continuously in the real world without external fine-tuning steps, while uman brains update themselves every moment through lived experience, without distinct training phases. While LLMs use large-scale unsupervised pretraining, their architecture is designed by humans with carefully curated fine-tuning strategies.
Humans learn language without needing structured datasets and token probabilities — just by hearing and experiencing the world. Machines simulate learning, humans experience it. The difference isn’t just scale, but nature.
Compute the operation of that feeling.
In other words, computing that feeling is equally mysterious whether it is done by neurons, or by transistors.
[1] There are attempts, like vague implications it has something to do with information processing - but that is not actually defining what it is, just what it is associated with and how it might arise. There are other problems with these attempts, such as the fact that the weather can be thought of as an "information processing" system, reacting to changes in pressure and humidity and temperature... so is it conscious? But that is tangential.