I think three of us have total conviction about the scaling law. That that I think we do. I do think the exact architecture choices and data mixtures is where the the devils are in the details.
But keep those methods you use to investigate things and monitor things very separate from the methods you use to generate reward which is something that AI companies including open AAI have held up as a principle especially in the case of avoiding putting training pressure on the chain of thought. So you might have monitors that read the agents chain of thought in order to alert you if something is going wrong somewhere but you don't train the agents with the outputs of that monitor.
It's like if I put electrodes in M1, you will probably be using a computer in an hour. Um and so it's just it's easier, it's more amenable to biology in many ways.
If you are working on material science, if you are working on pharmaceuticals, if you are working on chemistry, you in 2 3 years or even today are doing your role as a scientist dramatically different because of artificial intelligence.
open models are still incredibly useful, but fill a long-tail ecosystem relative to the closed counterparts that have monopoly ownership stakes in the most valuable areas like knowledge work collaboration, drug discovery, SWE, etc.
I think we're we're ready for complete re rewriting of how scientific discovery can be done because for ages, I don't even know how long, it relies on smart humans retaining what they have learned from other smart humans
the the breakthrough for us was realizing that AlexNet was not AlexNet. That AlexNet was an approach with deep deep learning that allows you to learn any function.
That is mostly in the realm of chemistry and is intrinsically far more tractable and predictable; given a validated target, good teams usually reach a decent molecule.
And there was no food or you know olive oil or food group or you eating more or less meat that changed the relationship between the Mediterranean diet and dementia risk. So what that tells me is it's much more about whole foods.
This took some special care, in particular avoiding any C++ standard library I/O functionality. Current frontier AI cannot handle this detail on their own.
My one little bit, the one little bit of of fundamental uncertainty even on long time scales is this thing about tasks that aren't verifiable. Like, planning a mission to Mars, like, uh you know, doing some fundamental scientific discovery like like CRISPR, like, you know, writing a writing a novel. Hard to hard to verify those tasks.
Teresa Torres is the best in the business at helping teams build products and services that their customers want. Here she shares her techniques for transforming your process into one of continuous discovery and learning.
the delusion comes from the reality that if you aggregated these breakeven revenues across companies, the market is not big enough to sustain all of them
so it's not a simple matter of programming it's not you know typing into a keyboard until or or I guess asking cursor to do something until a piece of software emerges, it's exploration, it's discovery, it's posing hypotheses and validating or invalidating those again and again and again.
It’s when you really don’t want to do a thing that you imagine a thing you don’t want to do more that’s worse than that and then in that way, you procrastinate by not doing the thing that’s worse. It’s a nice hack, it actually works.
If you notice that something repulses you and you have like a revulsion response to something, steer into it to learn more about it instead of avoiding it.
I would find it impossible to not pursue a discovery that I could imagine my way through, if I can really see how to get there. I cannot imagine abandoning it for some other reason, fear that it would be misused, which is a real fear.
So, I think she comes back in to relevancy in a different way than Friedman does, because I think in some ways she's tapped into a more universal human longing for independence and autonomy and self-creation and self-discovery.
Luis and Walter Alvarez, who made that incredible discovery, initially their discovery was based entirely on impact proxies, just as the Younger Dryas is. There was no crater. And for a long time they were disbelieved because they couldn’t produce a crater.
Another challenge for Class II biotechs is that even if you have a nifty computational drug discovery platform, you are still often bottlenecked by the rest of the drug development process. It takes 1,000 (or more!) small steps to advance a drug through discovery, development, and past the regulators.
For now, I am not convinced that issuing press releases about your compounds that talk about their discovery through AI techniques is sufficient to expect greater things from them.
I think when a system can significantly increase the rate of scientific discovery in the world, that’s a huge deal. I believe that most real economic growth comes from scientific and technological progress.
Such prolificacy is unlikely to be repeated any time soon; if current trends hold, a drug discovery scientist starting their career today is likely to retire without ever having worked on a single drug that makes it to market.
The most significant factor preventing us from taking better molecules into clinical trials is not our ability to design arbitrary molecules, but our (lack of) understanding of underlying disease biology. In other words, we often know how to make something, but not what to make.
In reality, the productivity crisis in the pharmaceutical industry is the culmination of decades of just about every aspect of drug discovery and development getting gradually harder and more expensive.
It's interesting to me that large language models in their current form are not inventions, they're discoveries. The telescope was an invention, but looking through it at Jupiter, knowing that it had moons, was a discovery.
I’d say for humanity, it’s both a tribute to the ability of discovery and the ability of really believing in things so that you have the confidence to go look for them, but it’s also a cautionary tale that you don’t want to assume things before they’ve been actually found.
A korrent is a belief a person has stated in their own words: one
sentence stating the claim, backed by a quote and a source, kept at
korrents.com.
Under a name here, the quoted block is what they actually said.
The korrent beneath it is the claim those words support, in
korrents' wording — tap it to see the record, its source, and who
else holds it.
Nobody here wrote their own korrents. They are compiled from public
statements, and a person can change their mind, which is recorded too.
About the English under a post
Some people here publish in a language other than English. Where they
do, this site shows a machine translation beneath the post, in
this typeface — the site's own, not theirs.
The post itself is never changed, moved or hidden: what is set in the
serif above is exactly what the person published, and it is what to
quote them on. A translation can be wrong in ways that matter,
especially about tone.
Only the post's own words are translated. A quoted post, a linked
article and a belief on korrents.com
are left in their original language.