phreeza

There are also players working on accelerating later stages of drug development with AI, e.g. https://atlas.bio/ probably just a matter of time until the big AI labs also start working on this.

Wow that's bananas. I suspected the answer would skew towards solar but not by that much. To me that also substantially discredits the "bureaucracy makes nuclear expensive" argument, my impression is the Chinese at doing this kind of engineering efficiently.

> China is doing it, even while they export the solar panels that everybody is buying.

Would be interesting to know how much nuclear vs solar they are building per year domestically.

The glory of having bested a stranger oh the internet in a battle of whits is enough for me!

I'll take that bet? Room temperature semiconductors are very common.

Nice idea in principle but (in the case of private) too long a time horizon for typical corporations to care about, sadly.

I agree with you there are other much grander challenges, but apart from that I don't think the list is horrible, because as I said it's incentive-aligned with the current moment we are in. Your suggestion would definitely be a good addition to the list for sure.

I think some of the items on the list are also way less likely to be worked on (like cryogenics) than others, because they really require some heavy infrastructure to iterate. My money for which of these gets cracked with help of an AI would be rubisco, that can be quite effectively worked on in a closed loop lab fashion with a standard lab or CRO.

Those don't have the property of being easily verified, which the problems proposed here do. I think that's a smart idea because it's a way to capture quick PR seeking AI-lab dollars for quite useful outcomes.

Next also include sales tax in label prices.

I agree, it is easy to miss just because we get used to things quickly. I remember this thought hitting me vividly when I saw this picture during COVID https://cdn.theatlantic.com/thumbor/MUBajBy-350E5JAjtntEhAYz...

Funny there the date is April 1985, a lot earlier and almost my birthday!

I think a lot of LLM progress in the last 2 years or so is really from densifying the training data to niches like this and not so much from general problem solving capability. I guess the hope is that the deluge of super specific training data will pay off through better overall competence. cf https://www.dwarkesh.com/p/the-sample-efficiency-black-hole

This is a pretty stunning result. The time series looks really convincing. Is the way the detector itself is trained orthogonal to this or could there be some "leakage" in that the pre-chatgpt text is in the (positive) training data?

For Switzerland, that's only true for the special tourist train panoramic carriages on the route, there are normal trains running on the same routes, too, that don't require reservations.

Or solar geoengineering? Which is scary but becoming more likely it seems.

I think this is a complete misunderstanding of what people mean by taste in software engineering. Taste is more like the System 1 response one builds to code over time, which (ideally) captures the quality of the software beyond surface level, so things like maintainability, composability, readability, likelihood of hidden bugs. This is completely different from the question if the code fulfills the immediate task at hand, but also not the same as pure aesthetics.

Was just going to ask if anyone remembered the name of this one. It was amazing! 15 year old me felt very important/futuristic syncing the news before school in the morning and reading during school break!

Edit: after looking it up with plucker as a starting point I think the one I actually used was AvantGo, same idea though I think.

Maybe its more like the hash was a well known secure hash but someone managed to extract the salt/private key/signing certificate from the camera?

I ran Claude Code on my ca 2015 ThinkPad which was having wifi issues and asked it to fix them. It diagnosed the problem and applied some obscure kernel flag which fixed the issue.

There are plenty of expenses in this order of magnitude that are not tied to direct increases in productivity. I think it may become a serious hiring impediment for companies to be really skimpy on these budgets for example.

Is it really that different with the current iteration of AI compared to what was possible 10 years ago? There may be some new awareness at the executive level of what is possible, but I feel like a "slacker detector" or whatever would have been possible with xgboost or lstms.

Pretty sure op is talking about university in Spain for other subjects, not Spanish class at university.

I think it's just a joke/reference to HAL in 2001: A Space Odyssey

The most amazing thing to me about Cider-V was that Cider (without the V) actually went away after a relatively short amount of time, when virtually every other internal service that is officially EOL-ed lives on essentially forever.

A rare breed!

That seems possible for generating completely new proteins.

Do you think it's also the case for lead optimization where you typically have some degree of measurements around your starting point, and you are expecting to stay in that local neighborhood for the generated candidates, too?

(Disclaimer: former Cradle employee here)

Check out manifold markets, sounds like that is what you are looking for?

Would you be open to sharing a version of your pitch deck? The main question in my mind is what kind of exit the VCs have in mind when they give you this money.

It seems like college towns have a lot of players even if they are small, what's the reason for that?

That's definitely true for some of them, but for others it's not so clear, like the Apollo or Manhattan projects? Those of course also have lasting impact but it's more in terms of knowledge, which at least arguably we are also accruing with these data centers.