- cross-posted to:
- technology@lemmy.zip
- technology@lemmy.world
- science@lemmy.world
- cross-posted to:
- technology@lemmy.zip
- technology@lemmy.world
- science@lemmy.world
We are constantly fed a version of AI that looks, sounds and acts suspiciously like us. It speaks in polished sentences, mimics emotions, expresses curiosity, claims to feel compassion, even dabbles in what it calls creativity.
But what we call AI today is nothing more than a statistical machine: a digital parrot regurgitating patterns mined from oceans of human data (the situation hasnāt changed much since it was discussed here five years ago). When it writes an answer to a question, it literally just guesses which letter and word will come next in a sequence ā based on the data itās been trained on.
This means AI has no understanding. No consciousness. No knowledge in any real, human sense. Just pure probability-driven, engineered brilliance ā nothing more, and nothing less.
So why is a real āthinkingā AI likely impossible? Because itās bodiless. It has no senses, no flesh, no nerves, no pain, no pleasure. It doesnāt hunger, desire or fear. And because there is no cognition ā not a shred ā thereās a fundamental gap between the data it consumes (data born out of human feelings and experience) and what it can do with them.
Philosopher David Chalmers calls the mysterious mechanism underlying the relationship between our physical body and consciousness the āhard problem of consciousnessā. Eminent scientists have recently hypothesised that consciousness actually emerges from the integration of internal, mental states with sensory representations (such as changes in heart rate, sweating and much more).
Given the paramount importance of the human senses and emotion for consciousness to āhappenā, there is a profound and probably irreconcilable disconnect between general AI, the machine, and consciousness, a human phenomenon.
What? No.
Chatbots canāt think because they literally arenāt designed to think. If you somehow gave a chatbot a body it would be just as mindless because itās just a probability engine.
Exactly. People see āAIā and think LLMs and diffusion models. Those are both probabilistic translation engines. Theyāre no more intelligent than an AC/DC converter, just a lot more complex.
However, there are neural networks and sense arrays in the field of AI, and those are designed to replicate the process of thought.
The real route to a thinking AI is likely a combination of the two, where a neural network can call on expert systems including translation engines to do the heavy lifting and then run a more nuanced decision tree over the results.
Thing is, modern LLMs and diffusion models are already more complex than a single human mind can fully comprehend, so we default to internally labelling them as either ālike usā or āmagicā, even when we theoretically know them to be nothing but really deep predictive models.
The problem is in the definition of intelligence.
To me, intelligence is simply problem-solving ability. It does not necessarily imply consciousness, having self-awareness or anything like that. A simple calculator is already displaying intelligence, even if limited to a very narrow situational set of problems, in the sense that it can resolve mathematical questions.
That doesnāt mean the calculator is self awareā¦ it just means it can resolve problems. Biological systems can also resolve problems without necessarily being aware of what they are doingā¦ does the fungus actually knows itās solving a maze the scientists prepared for it when it just expands following what is preprogrammed by its biological instincts determined by natural selection? Do the ants really know what they are doing when they find the shortest path just by instinctively following a scent of pheromones left by other ants?
Knowing exactly what causes consciousness is an entirely different problemā¦ and itās one that has not been resolved by any scientist or philosopher in a satisfactory manner. So we simply do not know that.
Seems to me your definition of intelligence ignores whole aspects of true intelligence, at least of the human kind, such as emotional intelligence and social intelligence and artistic intelligence and moral intelligenceā¦
āProblem solvingā is the name for what you described and it doesnāt necessarily require intelligence. In fact most intelligent people have encountered situations where it made solving a problem more difficult.
Yes there there as many types of intelligence as there are types of problems. Emotional intelligence deals with emotional problems, social intelligence deals with social problems. This doesnāt conflict with my definition, itās still problem solving.
Just because a being is intelligent does not mean it can solve all the problems of all kinds, it would require general intelligence, and even a generally intelligent being needs the right trainingā¦ if you are trained wrong or trained for a different kind of problem that does not fit the current one then your current experience might actually get in the way, as you point out.
Slime mold can solve mazes.
Yes, thatās what I meant 2 comments above by āfungusā (though to be fair, whether slime molds are fungi depends on your definition, they used to be classified as one, before āprotist kingdomā was made up to mix protozoa, algae & molds, but I keep preferring the traditional autotroph / absorptive heterotroph / digestive heterotroph division).
I also mentioned ants who can find the optimal path by simply following scents left by other ants without understanding how this helps with that.
You can be intelligent without being aware of your intelligence, or you can be stupid without being aware of your stupidityā¦ like how humans are actually creating problems for themselves in many cases.
Intelligence != awareness
If your definition of intelligence doesnāt include awareness itās not very useful.
I donāt know, I feel itās actually the opposite. Awareness is something you can only experience subjectively, itās āqualiaā, a quality that you cannot measure outside of yourself or detect externally. Thereās a reason IQ (āintelligenceā quotient) tests use puzzles/problems and donāt test conscious awareness. Most of the time in science intelligence is defined as problem solving and capacity to adapt/extrapolate because that definition makes it observable and more scientifically useful.
If it were to include awareness then we canāt in good faith answer the question: āis it intelligent?ā ā¦we can only say we donāt know. This is the main struggle of philosophy of the mind, what is often called āthe hard problem of consciousnessā. Empirical analysis would not show if something is having (or not) the conscious experience of being aware.
All the evidence suggests that our own minds are also nothing more than probability engines. The reason we consider humans to be intelligent is because our brains learn to model the events in the physical world that are fed into our brains by the nervous system. The whole purpose of a brain is to try and keep the body in a state of homeostasis. Thatās the basis for our volition. The brain gets data about about the state of the organism, and interprets it as hunger, pain, fear, and so on. Then it uses its internal world model to figure out actions that will put the body into a more desirable state. From this perspective, embodiment would indeed be a necessary component of human style intelligence.
While LLMs on their own are unlikely to provide a sufficient basis for a reasoning system, its not strictly impossible that a model trained on sensory data from a robot body it inhabits wouldnāt be able to build a representation of the world and its body that could be used as the basis for decision making and volition.
My understanding is that the reason LLMs struggle with solving math and logic problems is that those have certain answers, not probabilistic ones. That seems pretty fundamentally different from humans! In fact, we have a tendency to assign too much certainty to things which are actually probabilistic, which leads to its own reasoning errors. But we can also correctly identify actual truth, prove it through induction and deduction, and then hold onto that truth forever and use it to learn even more things.
We certainly do probabilistic reasoning, but we also do axiomatic reasoning i.e. more than probability engines.
I suspect that something like LLMs is part of our toolkit, but I agree that this canāt be the whole picture. Ideas like neurosymbolic AI might be on the right track here. The idea here is to leverage LLMs at parsing and classifying noisy input data, which theyāre good at, then use a symbolic logic engine to operate on the classified data. Something along these lines is much more likely to produce genuine intelligence. Weāre still in very early stages of both understanding how the brain works and figuring out how to implement artificial reasoning.
This completely understates the gulf between what we call AI and how the human brain actually works. The difference is so severe that acting as if theyāre quantitatively comparable is basically pseudoscience. You might as well start claiming that weāre not far off from building a Dyson sphere just because we invented solar panels.
Most āAIā these days are built using linear feed forward networks. The brain is constructed using nonlinear recurrent networks which are can do far more with less. Now you could theoretically create the same output from a linear feed forward network but itās way less efficient and would require many more neurons to achieve such a result. Which is wild when you consider that there are orders of magnitude more synapses in just the regions of the brain associated with language than there are parameters used in even todayās most advanced āAIā models. Now consider that human synapses rely on over a hundred qualitatively different neurotransmitters and not just a single 16-bit number. Itās also not just the scale of the signal that transmits information in a human synapse but the pattern too. Would you be surprised to know that there are a whole variety of signaling patterns neurons use? Because thatās true too. I havenāt even gotten into the differences in complexity in terms of how neurons process the information they receive. As of now there is no āAIā system that comes anywhere close to replicating that kind of complexity. Itās absurd to suggest where dealing with qualitatively similar machines here.
Way to completely misrepresent what I was actually saying. Nowhere was I suggesting that there isnāt a huge difference between the two. What I pointed out is that, while undeniably more complex, our brains appear to work on similar principles.
My only point was that the feedback loop from embodiment creates the basis for volition, and that what we call intelligence is our ability to create internal models of the world that we use for decision making. So, this is likely a prerequisite for any artificial system that has any meaningful intelligence.
Maybe try engaging with that instead of writing a wall of text arguing with a straw man.
Sure in the same way that a horse and a motorcycle operate on similar principles and serve the same function.
Where the straw man? Youāve missed my point entirely. LLMs and the human mind operate on categorically different principles. All the verbiage used to describe neural network models has little to do with how the brain actually works. Thatās honestly wasnāt a problem until Tech companies started purposely misusing those terms and now far too many people seem to think āAIā is something itās not.
A bold statement given that we donāt actually understand how the brain operates exactly and what algorithms that would translate into.
The straw man is you continuing to argue against equating LLMs with the functioning of the brain, something I never said here.
You appear to be conflating the implementation details of how the brain works with the what itās doing in a semantic sense. There is zero evidence that all the complexity of the brain is inherent to the way our reasoning functions. Again, we donāt have a full understanding of how the brain accomplishes tasks like reasoning. It may be a lot more complex than what LLMs do, or it may not be. We do not know.
Finally, none of this has anything to do with the point I was actually making which is regarding embodiment. You decided to ignore that to focus on braying about tech companies and LLMs instead.
Iām not claiming you ever said they functioned exactly the same way. Im simply stating that youāre way off base when you claim that they appear to operate using the same principles or that all evidence suggests the human mind is nothing more than a probability machine. Thatās not a straw man. You literally said those things.
Youāre betraying your own ignorance about neuroscience. The complexity of the brain is absolutely linked with its ability to reason and we have plenty of evidence to show that. The evolutionary process does not just create needless complexity if there is a more efficient path.
This is such a silly statement especially when youāve been claiming that both the brain and AI appear to work using the same principles. If you truly believe the mind is such a mystery then stop making that claim.
I donāt really care about your arguments concerning embodiment because theyāre so beside the point when you just blowing right by the most basic principles of neuroscience.
I bring up tech companies because theyāve had a massively distorting effect on how many computer scientists think the world works. Youāre not immune to it either simply because youāre a critic of capitalism. A ruthless criticism of that exists includes the very researchers whose work youāre taking at face value.
I literally said these things, and you never gave any actual counter argument to either of them.
Youāre betraying your ignorance of how biology works and illustrating that you have absolutely no business debating this subject. Efficiency is not the primary fitness function for evolution, itās survivability. And that means having a lot of redundancy baked into the system. Hereās a concrete example for you of just how much of the brain isnāt actually essential for normal day to day function. https://www.rifters.com/crawl/?p=6116
Thereās nothing silly in stating that the underlying principles are similar, but we donāt understand a lot of the mechanics of the brain. If you truly canāt understand such basic things thereās little point trying to have a meaningful discussion.
Thatās literally the whole context for this thread, it just doesnāt fit with the straw man you want to argue about.
Whose work am I taking at face value specifically? Youāre just spewing nonsense here without engaging with anything Iām saying.
Have some humility and willingness to learn.
I didnāt say it was the primary function. I guess all that talk about straw men was just projection. You donāt trust me, fine. Then what about Darwin who literally said, āNatural selection is continually trying to economize every part of the organization.ā Now please go and read some introductory texts on biology before trying to explain to me why Darwin is wrong. Thereās so much going on when it comes to the thermodynamics of living systems and youāre clearly not ready to have a conversation about it.
Youāre baseless assuming that hydrocephalus causes the brain to lose a substantial amount of its complexity. Where is the evidence for that? In most of these cases it seems much of the outer layers of the cerebral cortex are in tact. Itās also really telling that your citationās first source is an article titled āIs Your Brain Really Necessaryā which is followed in the Journal by another article entitled āMath and Sex: Are Girls Born with Less Ability?ā. But hey neuroscience hasnāt really advanced at all since 1980 right? The brain is totally redundant right? Thereās no possible way a critical and discerning person such as yourself could have been taken in by junk science, right?!!
I took issue with specific statements you made that stand apart from the rest of your comment. Thatās not a straw man. Although honestly this is on me. What can I expect from someone who thinks LLMs and the Human Brain are operating on similar principles? Youāre so wound up in a pseudoscientific fiction thereās nothing I can do. You might as well start believing in the astrology, crystals, and energy healing. At least those interests will make you seem fun and quirky instead of just an over confident tech bro.