4 ms·
> But nobody knows how GPT works. They know how it was trained, because the training scheme was designed by humans, but the algorithm that is executed during in
by smitty1e 4y ago
> But nobody knows how GPT works. They know how it was trained, because the training scheme was designed by humans, but the algorithm that is executed during inference was not intelligently designed but evolved, and it is implicit in the structure of the network, and interpretability has yet to mature to the point where we can draw a symbolic, abstract, human-readable program out of a sea of weights.
I object. ChatGPT executes in computer logic and is ultimately electrical signals in gates representing 1 and 0.
ChatGPT is vast and impressive, sure. Emergence[1] may get it past a Turing Test, fine. But it remains discrete logic.
In contrast, natural intelligence has not been reproduced organically, much less, fully understood. There is no repeatable experiment going from inorganic matter to self-aware, self-replicating life.
In summary, ChatGPT is impressive, but nowhere near capable of doing the impossible, e.g. predicting the weather with fidelity substantially into the future.
Nor can I bring myself to fret that Skynet is immanent.
[1] https://en.m.wikipedia.org/wiki/Emergence https://en.m.wikipedia.org/wiki/Emergence
- zetalyrae 4y agoI don't see the relevance. Natural intelligence has not been understood because if our brains were simple enough that we could understand them, we would be so simple we couldn't. This explains why it has not been reproduced. >There is no repeatable experiment going from inorganic matter to self-aware, self-replicating life. You can do the RNA world experiments in a lab, it would just take a lot of time and a lot of primordial soup, but eventually you would observe abiogenesis. I also don't see how abiogenesis and biology are relevant. >capable of doing the impossible, e.g. predicting the weather with fidelity substantially into the future. Thankfully this is not a measure of intelligence, since we can't do that either.
- glenstein 4y ago>Natural intelligence has not been understood because if our brains were simple enough that we could understand them, we would be so simple we couldn't. This explains why it has not been reproduced I don't believe that it works this way, for two reasons. One, the logic would seem to work the other way as well, if our brains are so sophisticated that they make natural intelligence possible, then we should be so intelligent that we would have the means to understand them. (At the end of the day I think this is a case where analogies aren't good enough to settle it one way or the other.) But secondly, the overall architecture of the brain can be understood in terms of underlying principles that are reapplied over and over, and so we can conceivably 'compress' the totality of information about the brain's architecture to the principles that explain why it functions. Granted we haven't done that yet and that could be extremely difficult, but, I don't think it's forbidden by necessity or by some transcendent principle. I would say that I agree both with your reply about RNA experiments being feasible, and I share your confusion as to how any of that was relevant.
- branko_d 4y ago> if our brains are so sophisticated that they make natural intelligence possible, then we should be so intelligent that we would have the means to understand them It may well be that building a brain is easier than understanding it.
- doctor_eval 4y agoSo what if it's electrical signals. It's not really 1s and 0s is it, they are grouped together into floating point numbers, right? And at some point there are enough ones and zeroes that you can effectively simulate just about any analog state. So why should we assume it's not possible to simulate the states of neurons? And why do we assume that the brain's structure is the only way to produce intelligence? Surely - like other biological systems - they are a specific implementation of what nature allows, but not necessarily the only one. > nowhere near capable of doing the impossible, e.g. predicting the weather with fidelity substantially into the future Why is it that people keep telling us GPT isn't capable of being conscious or having understanding because it's unable to perform tasks that individual humans can't do?
- smitty1e 4y agoWell, if we're to believe that thr Philosopher's Stone has arrived, why should it not have oracular power? I am regularly assured by the Climatistas that the apocalypse is at hand. What saith ChatGPT?
- doctor_eval 4y agoLiterally nobody is saying that about ChatGPT. Not sure what the climate has to do with this, other than maybe signalling whose team you're on.
- smitty1e 4y agoWell, the thrust of the headlines seems to be that ChatGPT is some sort of Delphic Oracle. Therefore, let it prophesy.