4 ms·
There's a long way to go from next-word-predicter to AGI. And that road might not ever end up in AGI.
by regularjack 3y ago
There's a long way to go from next-word-predicter to AGI. And that road might not ever end up in AGI.
- IshKebab 3y ago> next-word-predicter How tediously derisive. Part of the reason ChatGPT is seminal is that it shows that "next word predictor" can display extremely sophisticated capabilities; more than the sum of its parts. It might not end up in AGI but it's clearly a huge step closer - taking us from "will it every happen?" to "it actually might happen!".
- emptysongglass 3y agoIt's not a huge step closer. Consciousness, desire and will are orthogonal to ChatGPT's predictive modeling. Tedious is continuing to reiterate this point every time someone tries to spring the AGI doom-trap.
- jabradoodle 3y agoWe don't know that intelligence requires any of those characteristics, seems most likely it doesn't.
- ilaksh 3y agoThis thread demonstrates the central problem with these discussions which is that everyone has their own definition of AGI. GPT-4 clearly has a type of fairly general purpose intelligence. But there is more than one type of cognitive ability, some which LLMs can't have. The other person seems to consider AGI to mean something like "living digital person" or whatever.
- IshKebab 3y agoNope. AGI has a fairly well accepted definition - something with human level cognition. The ability to replace humans for most tasks. Bringing up things like sentience and consciousness muddies the waters because nobody knows those even are. HN is full of confident and unfounded assertions that ANNs fundamentally can't be conscious. Or even that they can't reason!
- ilaksh 3y agoWell, I agree that there is a reasonable and more correct definition of AGI something like you say, but unfortunately the term is not being used with that connotation consistently. The majority of people use it in a very fuzzy way, often incorporating ideas like consciousness. But also even in your statement, the word "level" is problematic. Because there is more than one dimension to the cognition. See Yann LeCun's recent criticisms, many of which are correct. Yet it's clear that LLMs have their own type of often useful and fairly general reasoning.
- IshKebab 3y agoCan you provide any evidence at all that they are orthogonal?
- emptysongglass 3y agoCan you? The burden of proof is on the AGI is nigh camp to prove they are not orthogonal otherwise it's just people handwaving and abusing the notion of "intelligence". I know what consciousness, desire and will are because I experience, and act, on them every day. People pointing at things outside themselves to prove intelligence have a much higher bar to clear. The very idea of intelligence as a discrete quantity is foolhardy and doesn't match what we know to be true in the animal kingdom. Intelligence is deeply rooted in an organism's sensorium.
- pixl97 3y agoConversely, AI will likely call you a 'next-meal-predictor' sooner than later.