I'm also skeptical that LLMs can ever invent new ideas.
I may be wrong about how soon the curve will flatten, and I may be wrong about LLMs fundamental limitations. But, I don't think it's extremely obvious that LLMs can have novel ideas or can grow into having novel ideas.
They're not there yet. Once they get there, that's literally the definition of Singularity.
But they are getting closer. Recursive Self-Improvement used to be a phrase people mocked LessWrong crowd for using and worrying about, now it's something both OpenAI and Anthropic already publicly admitted not only to pursue, but to already be benefiting from.
So far, I don't think the models are capable of running away on their own. Of course, it would be playing with fire to not at least consider the risks of such a runaway scenario and build in safeguards against it. But, there is no model that can build a better model on its own, thus far, to the best of my knowledge (which is far more limited than the models, so maybe I should ask them).
It starts with what they already claim to be doing - increasingly relying on existing models in non-trivial work related to training, evaluating and optimizing the next, more capable generation of models. As long as the proportion of work keeps shifting towards agents doing more and more of it, and humans less and less, that's RSI at play.
It may be that it turns out LLMs lack some fundamental level of judgement and it plateaus, but frankly I find this notion absurd; LLMs already show better judgement than most people. The alternative is, at some point LLMs will show the ability to futz their way into improvement of the next generation of models even without humans in the loop - even if much less efficient at first, if generation N+1 is more capable than generation N, it'll either take off or burn out.
just because more and more agents are doing human work, that in no way means the model somehow becomes magically more intelligent, it just means the work will stall and continue on at the same level forever
hell even if they hypothetically have an internal model that can output the entire training data set in a better format, there's no scientific evidence that the newer format has new information that is sufficient enough to train a better AI
as a matter of fact the scientific evidence is on the contrary
All intelligence, LLM or not, is bound to plateau around the point where the need to operate within physical reality bottlenecks the speed of feedback. AI is progressing swiftly in the digital realm where feedback is nearly instantaneous, but it's unclear whether that would translate into improvements in the physical world where signals are much noisier and intelligence and judgment are less impactful.
My go-to example: wheels suck at mobility in the natural world. There's a reason no animal uses wheels as their means of locomotion. They only really work on flat, hard surfaces that give a good grip.
Our solution? We didn't build mechanical legs for all-terrain mobility. We paved the world instead. Dirt roads, then brick roads, then asphalt and concrete - since ancient history, we were forcing the world to adapt, so the one means of mobility we could easily built was effective.
We repeat this pattern every time when the world doesn't agree with us, and adapting ourselves to it is too much of a hassle.
Now, AI will likely adopt the same approach to these kinds of problems, which may not end up very nicely for us.
(Note: we are already adapting our own digital worlds to make them easier for AI. Much like building roads for our wheels, we now have a resurgence of CLI, with new tools exposing high-level operations optimized for agents - not humans - to use comfortably.)
With AI it's a bit of a chicken or the egg situation. Before it can ablate the complexity of the world, it needs to be able to deal with it, including adversaries such as ourselves. For this it needs a tool as sophisticated as our bodies, but could AI develop such a tool for itself without the help of this tool? I'm not convinced.
I can already see the border shift even for mundane tasks I have Claude working on. Increasingly, I'm just setting a high-level goal, and then checking progress and occasionally answering questions or doing something like configuring a system Claude can't easily reach itself (e.g. recording a bunch of traces through my normal use of a system that Claude deemed too fragile to risk operating on its own). Of course, I get detailed instructions to help me - "go there, do this and that, then press this to capture recording, run through this script here to process, attach result to next message". In those cases, Claude is effectively using me as a tool to call.
weve seen some improvement from the LLMs unattended, maybe, but will it actually keep improving vs needing a human to bring it back on track?
the recursive part is that it keeps improving on itself, but we really have no example of that. if it does it 30 times with improvements, then maybe, but even then, to actually be relevant it has to do better than paying scientists to do the work for the same cost, consistently.
RSI still means nothing if it costs 1000x the cost to get the same improvements as a human researcher
This is the nuance that poster doesn’t understand. Given how much money thrown at it - we’re not even close. Who has the appetite to keep throwing more given they continually need to keep raising fresh money?
Money is fake and not a constraint, that's literally the point of capital investments you guys are overindexing on so badly.
> Could vehicle factories be more automated if we threw a gazillion dollars at it?
They already are automated as much as it makes sense. Some of the automation is silicon-and-steel based, some of it is protein-based. Car manufacturers aren't in the business of pushing robotics and nanotechnology, so they prefer to hire protein automatons instead of developing and building their own, so yeah, they "pay people", but think what exactly they are paying them for.
Then consider that this is very much the work the AI is gradually getting as good as, or better, than us.
> This is the nuance that poster doesn’t understand. Given how much money thrown at it - we’re not even close.
What you seem to be missing is that "investing in AI" isn't investing in a chatbot, it's investing in technology that will (and already partially is) sit upstream of every other industry, of everything humans do. Like electricity or the Internet itself.
(The other thing you seem to not understand, in contrast with some of the investors, is that RSI is not a linear walk, it's an exponential curve. X-risk notwithstanding, by the time it's obvious to everyone, it's too late to make money investing in it.)
> keep throwing more given they continually need to keep raising fresh money
Have you heard about R&D?
I'm starting to think that "investors first" thinking that's so common here, that makes people feel they're smart, is actually quite backwards, especially at this scale. Or maybe it's simply people starting with a conclusion and trying to fit reality to match it, no matter how clear of a nonsense that conclusion is?
Not necessary. Scientists are capacity limited and supply limited.
> RSI still means nothing if it costs 1000x the cost to get the same improvements as a human researcher
It means you can replace a human researcher with 1000x their salary burned on electricity. It also mean you can get two of them for 2000x of salary of one,
That's a bargain, actually, even with the anomalous, absurdly-overinflated salaries in top-tier ML.
I you knew it was consistent, then these companies would immediately fire their scientists, and burn 10 000x as much as mean researcher salary this month to be able to 10x their virtual headcount overnight, and then use that to make the 1000x be 500x, then 250x, then 125x, then ... and at that point they'd had all the money in the world, because even more skeptical investors would notice what's going on there.
--
TL;DR: what you all seem to miss is that electricity scales better than people.
At least they’re led by trustworthy and honest people or we’d need to take their claims with some dose of skepticism.
So far, that's not exactly how it's played out. Humans are still necessary for the leaps in capability or efficiency. A model can grind on a problem to eke out the most performance, and models can synthesize data and iterate on various techniques to find the optimal combination. But, seems like humans still have to provide the real thinking, and the talent and drive for doing that is not concentrated in one company or city or even one country. And, (surprisingly) a lot of the people involved are in it for advancing the field more than making another billion dollars, so they're publishing their research.
So, yeah, the moat isn't deep. Even the compute moat, that OpenAI, Musk, and a bunch of other also-rans (like Oracle) bet the farm on, isn't really panning out. The Chinese makers just spent their effort on making models vastly more efficient, since they couldn't do anything about having an order of magnitude less compute available.