31–40 of 51 posts
What's more alarming isn't that AI is limited to existing domain data, it's that when people push it to deviate outside those known data points it confidently hallucinates nonsense.
[flagged]
We know the pre-trained models do tend to revert to mean, but I don't think that's enough to say SFT / RL models will do the same, although some might argue RL only sharpens the distribution, even for that, I am skeptical about that paper.
[flagged]
[flagged]
My hot take is that mathematical and scientific 'soundness' is ultimately more of an aesthetic preference than an objective quality of reality. Good science makes sense to humans, and 'what makes sense' is ultimately what fits satisfyingly in your brain. There's nothing inherently wrong with an enormous epicycle model of reality from the perspective of the God of Math; so long as your formal system is consistent and…
My hot take is that mathematical and scientific 'soundness' is ultimately more of an aesthetic preference than an objective quality of reality. Good science makes sense to humans, and 'what makes sense' is ultimately what fits satisfyingly in your brain. There's nothing inherently wrong with an enormous epicycle model of reality from the perspective of the God of Math; so long as your formal system is consistent and…
If biology, or some other subject area, is inherently, irredeemably hard to explain, and always will be, then I don't care about it much, because it doesn't mean very much. I care about explanations, not "reality" in the sense of every arbitrary muddle of knotted nerve fibers and confused flour beetles. If all the world's messy, inexplicable things were to gang up and cause us trouble such that we have to pay them at…
Please don't editorialize titles unless they're clearly clickbait. "Designing AI for Disruptive Science" is a bit market-ey, but "AI Risks 'Hypernormal' Science" is just a trimmed section heading "Current AI Training Risks Hypernormal Science".
Earlier quoted context omitted.
I don't think paradigm shifts have to be 'better' in some march-toward-progress sense, they can be lateral or even regressive in that way and still lead to longer-horizon improvements. I think also what's practically applicable changes constantly. Perhaps we're truly at the End of Science, but empirically we've been wrong every other time we've said that. My money is that there's more race to run.
> I don't think paradigm shifts have to be 'better' But they do. Paradigm shifts happen because the new paradigm explains the unexplained and importantly also covers the old model. If prior data is unexplained with a paradigm shift, the shift will never be adopted. > Perhaps we're truly at the End of Science Who said that? Just because the core of our current models seem pretty rock steady doesn't mean there's not mo…
Empirically it seems that paradigm shifts are more driven by deaths and retirement rather than improved fit to the data. Moreover the way that you reconcile old data with the new model can be contestable; it's not like everyone all at once says "oh this new model is clearly a strict superset of the previous one, time to adopt it". With all that said I think one could argue that this stuff is basically noise and that the process still 'trends toward progress' (and I'd agree). But I would say that the scale of noise can also be quite large relative to things a human might experience in their life. I was sort of imagining social-disruption (like a dark-age type regression) as the 'backwards paradigm shift'.
> but there's likely not going to be major "relativity" moments from here on out
I cannot understand how anyone treat this as something that can be objectively concluded; by definition these kinds of radical paradigm shifts are basically unforeseeable up until they happen. I called it the "End of Science" to draw a parallel to "End of History"-type thinking because both (IMO) take this view of "there will be no more revolutions, only incremental adjustments on an unshakeable core into infinity", which I feel is personally a 'vibes based' assessment of things. It's not even that I disagree with it so much as I feel like the statement is basically (and will always be) a pure guess, one which many people have made and been wrong about in the past.
The article presumes that the models we have today describing everything could still be subject to a major paradigm shift. Maybe they could be, but it seems pretty unlikely. The edges of a lot of scientific understanding are now past practical applicability. The edges are essentially models of things impossible to test. In fact, relativity was only recently fully backed up with experimental data.
Please don't editorialize titles unless they're clearly clickbait. "Designing AI for Disruptive Science" is a bit market-ey, but "AI Risks 'Hypernormal' Science" is just a trimmed section heading "Current AI Training Risks Hypernormal Science".
Thanks, but please email us (hn@ycombinator.com) when you see things like this, so we can take action more quickly.