Is the brain a biological computer? \ stacker news ~AskSN

pull down to refresh

2345 sats \ 32 comments \ @k00b 10 Jun AskSN

Ilya Sutskever¹, one of the founders and chief scientist of OpenAI, gave a commencement speech at University of Toronto recently. He encourages the graduating class to take AI seriously, because:

We have a brain. The brain is a biological computer so why can't a digital computer, a digital brain, do the same things?

The main objection to this, at least among ones that I've heard, suggest that there's an immaterial component to the brain, a soul or a source or what have you, that isn't biological nor computer-like. Another objection that I've heard from Chomsky is that the brain has parts that are not exactly computer-like, or at least, they aren't simply an issue of scaling training data.

I'm curious what stackers think.

I really enjoyed this interview with Ilya from a few years ago: #324711 ↩

view all related items

121 sats \ 0 replies \ @nullama 11 Jun

We have a brain.

Yes.

The brain is a biological computer

Not necessarily in the same way we know computers. We still don't really fully understand how the brains works.

I will leave this quote here:

All computers are information-processing systems, but not all information-processing systems are computers, and the human mind is an information-processing system that is not a computer.

This article explains the point further, focusing on the views from Alan Turing, the father of theoretical computer science.

147 sats \ 4 replies \ @Scoresby 10 Jun

In my mind, computer implies determinism: if you know the inputs, you know the outputs.

I believe human brain (somehow) cannot be predicted. Human brains can produce surprising outputs from known inputs.

I don't think it's all just a matter of complexity.

102 sats \ 2 replies \ @elvismercury 10 Jun

Just because you personally can't predict it doesn't mean the logical structure for predictability is not in place.

0 sats \ 1 reply \ @Scoresby 10 Jun

Well, sure, but I was trying to refer to the impossibility of prediction by anyone.

I'm realizing prediction is not the right word. Determined by is more what I mean, but it is clunkier.

A computer will have certain circuits light up because of certain inputs. Neural circuits in a human brain may light up because of certain inputs...or they might not.

Sure, I may not be able to perceive the true logical structure of human brains, but I may choose to reason about them anyway, or I may not. I'm not sure this is true of a computer.

17 sats \ 0 replies \ @elvismercury 10 Jun

The circuits that light up do so stochastically as a function of inputs. If you clamp these, you get neural activation that is not mysterious, even if unknowable by you (or anyone). That's the heart of the problem with the "unknowable" critique.

102 sats \ 0 replies \ @Undisciplined 10 Jun

I thought Brett Weinstein had a good point about this. Consciousness presumably carries a cost, as most adaptations do. So, it must do something to justify that cost.

The most obvious thing it might be doing is making better decisions, which implies that we’re not fully predictable.

That’s a fairly condensed version of his argument, but I think you can see what he’s driving at.

97 sats \ 1 reply \ @Aardvark 10 Jun

Reminds me of that old question, does free will actually exist? I don't know, maybe I'm just a meat robot running on instincts and assuming I'm the one behind the wheel. Given how terrible the human brain is at storing information, I assume most of the things I remember didn't even happen, or definitely not the way I remember them happening.

Either way, I wouldn't mind a firmware update, so here's to hoping.

42 sats \ 0 replies \ @Undisciplined 10 Jun

You probably wouldn’t be sentient if that were the case.

92 sats \ 1 reply \ @freetx 10 Jun

No. The brain is a quantum antenna (ie. the Penrose theorem).

Your brain, like all your other organs are passive sensory. Your tongue taste, your eyes receive light, etc. Your brain detects consciousness waves like a radio....

21 sats \ 0 replies \ @Bishop 11 Jun

I felt compelled to look up the Penrose theorem....

Perplexity:

The idea that the brain functions as a "quantum antenna" stems from the Penrose-Hameroff theory of consciousness, also known as Orchestrated Objective Reduction (Orch OR). This controversial hypothesis proposes that quantum mechanical processes within neurons—specifically in structures called microtubules—enable consciousness by interacting with fundamental spacetime geometry. Here's a breakdown:

1. Quantum Processes in Microtubules

Microtubules are cylindrical protein structures in neurons that help maintain cellular shape and regulate synaptic activity. According to Penrose and Hameroff:

Quantum superposition occurs in microtubules, where tubulin proteins (their building blocks) exist in multiple states simultaneously[1][3][6].
These quantum states are shielded from environmental interference (decoherence) due to the microtubules' lattice structure and isolation within neurons[4][6].
Quantum coherence—a synchronized quantum state—could span large neural networks via gap junctions (electrical connections between neurons)[7][8].

2. Penrose's Objective Reduction (OR)

Penrose argues that classical physics alone cannot explain consciousness. His theory introduces:

Quantum gravity threshold: Superpositions in microtubules persist until reaching a critical mass-energy separation, causing spontaneous collapse (objective reduction). This collapse is governed by spacetime geometry at the Planck scale[3][6][8].
Non-computable influence: The collapse outcome isn’t random or deterministic but influenced by "protoconscious" properties embedded in spacetime itself[3][6]. This links consciousness to fundamental physics.

3. The "Quantum Antenna" Analogy

The term "quantum antenna" metaphorically describes how microtubules might:

Detect/transmit quantum information: By leveraging quantum coherence, microtubules could process information beyond classical limits, acting as receivers or transmitters of quantum-level phenomena[4][6].
Interface with spacetime geometry: Penrose suggests OR events reconfigure spacetime at microscopic scales, implying the brain interacts with universal quantum fields[6][8].

4. Criticisms and Challenges

Decoherence issues: Critics argue the brain’s warm, wet environment would disrupt quantum states too quickly for meaningful computation[2][7].
Lack of empirical evidence: No experiments have conclusively demonstrated quantum processes in microtubules affecting consciousness[7].
Alternative theories: Mainstream neuroscience attributes consciousness to classical neural networks, not quantum effects[5][7].

In summary, the "quantum antenna" concept posits that the brain’s microtubules harness quantum mechanics to access non-computable aspects of reality, with consciousness arising from spacetime-level interactions. While innovative, the theory remains speculative and contentious within the scientific community[2][5][7].

[1] https://www.europeandissemination.eu/can-consciousness-be-explained-by-quantum-physics/14424 [2] https://scientificandmedical.net/roger-penrose-on-consciousness/ [3] https://royalsocietypublishing.org/doi/10.1098/rsta.1998.0254 [4] https://nautil.us/roger-penrose-on-why-consciousness-does-not-compute-236591/ [5] https://en.wikipedia.org/wiki/Quantum_mind [6] https://en.wikipedia.org/wiki/Orchestrated_objective_reduction [7] https://www.livescience.com/37807-brain-is-not-quantum-computer.html [8] https://legacy.cs.indiana.edu/classes/b629-sabr/QuantumComputationInBrainMicrotubules.pdf [9] https://mappingignorance.org/2015/06/17/on-the-quantum-theory-of-consciousness/ [10] https://academic.oup.com/book/44484/chapter/376471400 [11] https://www.quantamagazine.org/a-new-spin-on-the-quantum-brain-20161102/ [12] https://www.newscientist.com/article/mg25634130-100-roger-penrose-consciousness-must-be-beyond-computable-physics/ [13] https://www.sciencedirect.com/science/article/pii/S1571064513001188 [14] https://pubmed.ncbi.nlm.nih.gov/19792156/ [15] https://www.youtube.com/watch?v=xa2Kpkksf3k [16] https://arunsingha.in/2023/10/17/the-quantum-brain/ [17] https://en.wikipedia.org/wiki/The_Emperor's_New_Mind [18] https://physicsworld.com/a/quantum-theory-of-consciousness-put-in-doubt-by-underground-experiment/ [19] https://journals.sfu.ca/jnonlocality/index.php/jnonlocality/article/view/51

115 sats \ 0 replies \ @SimpleStacker 10 Jun

I'm a Christian so I certainly believe that there's a soul that can't be replicated by a computer. I think it's also safe to say that the soul isn't entirely contained in the brain, because the soul continues to exist even if the body is burned to ashes. As to whether or not a computer can fully replicate the functionality of a brain, I can't say.

100 sats \ 0 replies \ @fauxfoe 10 Jun

At some level of abstraction, sure. Brain is computer. But it's not an LLM. And it's not a CPU. The brain is a bunch of parts that integrate a constant stream of sensory input and the parts do a bunch of things that add up to a bunch of outputs. I guess there's some language stuff in there, but that's not the whole story at all, and it's not, AFAICT, predictive. And it also doesn't include all the stuff that happens outside the brain. Or in other systems.

So yeah: brain is computer. That's not entirely false. Just not sure how useful it is as an observation.

55 sats \ 0 replies \ @fm 10 Jun

Is the brain a biological computer?

I do think so. We are definitly emulating our own brain in computers..

Funny we also work with RAM and ROM kind of setup.

When one looks into the Fractal hypothesis end up question phrases like:

"Let us make man in our image, after our likeness..." Genesis 1:26

And as we like to play god..
Maybe the transiction from Carbon Based Life to Silice Based life is just one more step on evolution and we are the "gods" making the computer after our likeness.

21 sats \ 0 replies \ @0xbitcoiner 10 Jun

Interesting thought, but I feel like that comparison has always been a bit of a shortcut to explain the brain's ins and outs. It's a widely used metaphor. Personally, I'd describe the brain as a supercomputer that's still in the realm of fiction. Who knows if we'll ever truly get there, but that's the direction we're headed, isn't it?

25 sats \ 0 replies \ @grayruby 10 Jun

I think the analogy is fine. The brain functions in a similar fashion to a computer but it is also adaptive where a computer is not.

10 sats \ 0 replies \ @LibertasBR 10 Jun

Looks like gaslighting #1001568

116 sats \ 0 replies \ @kepford 10 Jun

What bugs me most about pop discussions on AI is how little critical thought seems to be applied. Now I know this is no different than any other topic but its so blatant how most people are being duped.

There are these stories about AI doing some scary thing in testing and zero mention that this all just happening in a chatbot. A bot that is designed to produce output that seems like a human.

Its not really thinking. Its calculating what you expect to hear
Its not actually able to do anything other than output text

So all the fear is on the machine side yet the danger is the same as with any technology. The people making it and using it. The average person doesn't understand this and I don't hear enough people calling this out and the dangers of this at scale.

I don't care that much that Scam Altman is pumping his bags. I care more about scaring people into surrendering more freedom to the state and the technocracy. I care more about people being black pilled. People thinking we need to just go to a UBI (socialism). That will kill massive numbers of people.

10 sats \ 0 replies \ @FmpPerspective 11 Jun

The flaw lies in assuming computation is the highest form of cognition. A computer processes; a mind intuits, reflects, creates meaning. Call the brain a biological computer if you must—but don’t forget: we love, we choose, we suffer. There’s more going on than logic gates.

10 sats \ 0 replies \ @didiplaywell 11 Jun

The fact the brain is a computer is not to be confused with its consciousness itself. Consciousness is not intelligence, so we have our computer brain on one side, and our consciousness on the other. Confusing both as a same thing leads to the pseudo metaphysical explanations, based on "soul" and immaterial means, that come from people that do not know what the brain nor a computer is.

50 sats \ 0 replies \ @fiatbad 10 Jun

This is a copy/paste of my old comment here: #864172 and here: #863850

Currently, we can prove with quantitative evidence that a neural-network can form at least the basics of intelligence. The more neurons in the network, the more intelligent and sentient-like the entity will seem.

Humans are (roughly) at the top of the list of animals for their neuron count: https://en.wikipedia.org/wiki/List_of_animals_by_number_of_neurons There are a few animals with neuron counts similar to ours, and I believe those animals could do everything a human could do and are every bit as sentient as we are.

In the list I linked, you see elephants are one of the only creatures with a higher neuron count than humans. Most likely, they have the mental capacity to build cars, planes, and eventually travel the stars. However, they lack other important biological features, unrelated to intelligence, that prevent them from doing so. For example, their size and the fact they have no hands.... makes it hard to create agricultural tools, or type on a keyboard.

So, yes, cetaceans and squirrels are sentient and intelligent. But their low neuron count make them less intelligent and less sentient. I believe intelligence and sentience scale exponentially with neuron count. As long as there are at least 2 neurons present, intelligence and sentience exists... albeit not to a significant degree. Modern day AI's have the neuron count of something like a dog probably. Would have to look into it more to compare them.

Nature is full of examples of many small things working together to create something which appears completely different. https://plato.stanford.edu/entries/properties-emergent/ It seems crazy that large organisms are made up of billions of small, independent cells. Such a thing would seem supernatural to the untrained eye. But it's just mathematical induction. If a simple eukaryote can combine four cells together in order to create something "greater", then why not combine 8 cells? 16... 32..... 512.... 1 Billion.

https://www.u-tokyo.ac.jp/focus/en/articles/a_00220.html

Google AI tells me that the average human body contains between 30 to 37 trillion cells. Tiny, individual things. By combining them together, the (seemingly supernatural) properties of animal and plant life emerges. We can use induction-like mathematics to prove that there is nothing supernatural about it. If 4 cells can come together, so can 8, so can 30 trillion. For organisms with less than 1 Million cells, not much is possible. Such organisms will never be capable of understanding mathematics, driving cars, or traveling the stars. There is a sweet spot, somewhere between 1 Million cells and 30 Trillion, whereby something supernatural-like emerges, endowing the larger celled organism to do something even greater than its cell count should allow. In other words, it's not a linear scale. The cell count may have gone up by a factor of 1,000, but because of the power of "emergent-properties" the potential for the organism went up by a factor of 10,000.

Sense we are able to use induction to prove that nothing supernatural was required in order to go from a 4-celled organism, to a 30 trillion-celled one, it would be silly for someone to attempt to add supernatural conjecture to the mix.

You might argue that it's impossible to quantify "emergent-properties", so that's where supernatural hypotheses are usually added. I would argue that most emergent-properties that were once nearly impossible for us to measure are about to become much easier to measure because of AI. Our AI's can keep track of the interactions between the trillions of variables in a given equation. Take the folding of protein as an example: https://www.science.org/doi/10.1126/science.370.6521.1144

Neurons in the brain are part of those 30 Trillion cells in the human body. All the same concepts apply to how consciousness is an emergent-property of having more neurons. A human brain contains around 80-90 Billion neurons. Compare a brain containing 100 neurons with a brain containing 85 Billion. Somewhere between 100 and 85B, emergent-properties arose which made the sum greater than the parts. Not because of anything supernatural or unmeasurable, but because of emergent properties and induction. If I can prove 4 neurons can work together to solve problems and make decisions for the whole, then I can also prove it works for 8 neurons.... 16.... 32..... 85 billion. And it's not linear because of the emergent-properties and work. A brain going from 100 neurons to 85 may be an increase of 1,000 times, but the practical functionality of that brain goes up more like 10,000 times.

"Then comes the attempts to describe these phenomenon with physical sciences. Again, no evidence."

Using induction, why couldn't a sufficient number of neurons be enough to create all kinds of crazy experiences for itself. Dreams, ideas, creativity, and even "visits to an unearthly dimension".... all within the realm of basic computing. Have you seem the videos of modern day, artificial neural-networks having "dreams"? Looks otherworldly to me. Near-Death-Experiences, to me, seem a lot like the average DMT experience. The compound makes the neurons in the brain configure in a way that is not normal, but still coherent enough to retain its visuals and experiences. It disconnects itself from the external world and focuses inward.... similar to what happens when we dream. And, sense all humans have similar brain structures and similar neuron counts, they experience similar NDE's and DMT trips.

Why do we need to throw anything supernatural into this? There is real science to show how neural-networks work. And with AI, we may be able to quantify the emergent-properties piece of consciousness and AI.

People who claim there must be a supernatural element are committing the God-of-the-Gaps fallacy: https://en.wikipedia.org/wiki/God_of_the_gaps#:~:text=The%20term%20God%2Dof%2Dthe,an%20argument%20from%20ignorance%20fallacy.

For all time, humans have been coming up with supernatural hypothesis for things we couldn't understand. Eventually, science could explain more and more of nature, and the supernatural began losing ground. Yet, even today, people continue to make this fallacy. As soon as we reach the limits of modern science, gods and other supernatural conjecture takes over.

Why don't people apply "Occam's razor" to everything in their lives? https://en.wikipedia.org/wiki/Occam%27s_razor Why the need for something greater? Why the need for something supernatural? Why not live with humility and the "don't trust, verify" ethos?

5 sats \ 0 replies \ @kepford 10 Jun

I agree with Chomsky, which is rare for me. My general thought is that people with an interest in pumping their bags will exaggerate the power of their products. So of course these guys are gonna say things like this. I'm not saying they don't believe them. They may or may not. The point is to keep their bias in mind.

What I hear people (smarter than me) say is that their is the brain and the mind. I'm not gonna pretend to understand either but my gut tells me that the AI's (different kinds) are something different from a organic brain. If you believe that the human brain evolved to be useful to humans over a massive amount of time it would not make sense to try to completely replicate it in a machine. Just as we do not try to replicate how humans lift things. Machines can lift far more but they do it better because they do NOT copy humans.

I think the dudes that pitch AI are using parlor tricks to impress humans into buying into their businesses. They are imitating human language and statistically predicting what we humans would expect to get as an answer. There are similarities with how humans think but its different. I'm not even saying this is bad that chatbots are doing it different. Its what I would expect a computer to do.

There are many issues with our brains and how we think. We have many deficiencies and many of us struggle with our brains or minds. Not sure which one actually. Maybe its both. I don't think these machines are trying to build those in nor should they.

We can see from history that you can make machines that do pretty much any repeatable task better than a human but its not a human. Chess is a common example and yet people prefer to watch a real human play the game.

This whole topic is interesting to me and what it continues to point out to me is how massive the gap is in our understanding of the brain and the mind. Its easy for us to use our cultural lens and be blind to how much we still do not know. While there have been massive leaps in understanding of our bodies we are still a long way from even understanding the brain and many ailments that affect our bodies. I firmly believe we are in the dark ages of medicine still. We still attack illness with blunt instruments. Don't get me wrong. Its massively better than it was 200 years ago but I think people don't get how far we have to go until they loss a family member to cancer or another common disease.

So it is VERY hard for me to believe that AI is anywhere close to an actual intelligence or a mind. I will admit that while my understanding of LLMs and chatbots is much better than it was a year ago I still do not have a really strong understanding of neural networks. On top of this I rarely hear anyone talking about AI that can explain it either. So I'm left thinking many people working on these machines are still trying to understand how they work... which isn't bullish on a massive breakthrough IMO. Then you top off the massive inefficiency for running these machines vs the value they provide and I think we are in for a massive correction in the next 5-10 years. That could set AI back for a long time.

30 sats \ 0 replies \ @pillar 10 Jun

I think looking at a baby laugh is proof enough that human brains and LLMs are two different things.

If you haven't watched it, you might enjoy this movie. Came to mind with your reference the brain's soul.

0 sats \ 0 replies \ @398ja 11 Jun

I think this is an analogy, similar to the "second brain" as a Zettelkasten (slip box).

I like to think of AI as a Zettelkasten on steroids, an external "brain" that extends memory, a very helpful thinking and creativity support, and much more...

0 sats \ 0 replies \ @nkmg1c_ventures 11 Jun

I thought that's what organoid computing is about

0 sats \ 0 replies \ @Gekido 11 Jun

Now I feel like a robot after this information

0 sats \ 0 replies \ @Solomonsatoshi 10 Jun

Yes, designed in large part by evolution to compute the myriad of variables required to function and survive best in a social group.

Humans are weak individually- but very capable of leveraging collective effort and networking.

Thus governments have an unavoidable role to play in human society and Libertarians are taking the notion of individual sovereignty too far in many cases.

0 sats \ 0 replies \ @guerratotal 10 Jun

The brain definitely shares similarities with a computer—processing input, storing information, adapting over time—but calling it a "biological computer" might oversimplify things. Consciousness, emotion, and subjective experience aren't things we can easily code or train. We’re still scratching the surface of how the brain actually works, and pretending it’s just a really advanced calculator might miss what makes it uniquely human.

0 sats \ 0 replies \ @denlillaapan 10 Jun

I mean, -issshhh

0 sats \ 0 replies \ @ken 10 Jun freebie

Hey kid

I'm a computer

0 sats \ 0 replies \ @BallLightning 10 Jun

Yes, of course. The architecture of the current AI systems is built on something called artificial neural network, which is a think that is somewhat similar (and inspired) from the biological neural networks, such as the human brain. Now if you want to replicate the behaviour of the human brain you will probably need to simulate the whole thing. And you will probably need to simulate even seemingly unrelated things like the metabolism and the blood supply. But theoretically if you do if, you will externally observe the same thing you wolud ovserve from a human.

0 sats \ 0 replies \ @itsrealfake 10 Jun

roger penrose & "microtubules"

0 sats \ 0 replies \ @RideandSmile 10 Jun

it reminds this video clip

Footnotes

1. Quantum Processes in Microtubules

2. Penrose's Objective Reduction (OR)

3. The "Quantum Antenna" Analogy

4. Criticisms and Challenges