1. It shows what even this wave of AI can actually do.
2. I wish it were done by different folks, ideally under some kind of public control like NASA research or the NPR model.
3. Keep in mind: natural science is different. It's not always a matter of computation. Computer science folks often struggle with this -- but this virtual world here does not actually exist. Everything is physical, including information. Any natural science PhD or otherwise knows just how complicated nature actually is -- e.g. mention any research topic and try to encapsulate all the relevant phenomena present there. Pure mathematics is different because we define the problem, rarher than explore nature. We are in my view far away from removing humans in natural science R&D. Advancements in AI however can greatly assist us in all natural sciences, which is already beginning to happen.
Most experimental physics and other natural sciences are strongly driven by their theoretical siblings, i.e. in particle research nothing gets built without a solid theoretical foundation of what you expect to find (or where you expect existing theories to break down), the same is true in other areas, no one is doing an experiment in quantum physics before they have a solid theoretical understanding of the effects they try to see. I think AI can come up with great experiments. And if epxeriments lead to results that are unexpected AI can help with that as well.
So I'm greatly excited what AI will bring about in physics, more so than in math, because in physics it's clear that our fundamental theories are missing a big piece of the picture, and given how easily AI crunches through Millenium prize problems I think it's possible that AI will come up with a viable grand unified theory uniting quantum mechanics and gravitation, or produce new predictions in other areas. There's enough contradictory or unexplained observational data available to make a ton of progress on the theory side I think. Exciting times ahead!
Agreed! Many people are saying AI isn't really intelligent yet because it can't come up with genuinely new things. Maybe finding a rigorous formulation of QFT / high-energy physics would be a great test for whether they are!
The problem with physics and chemistry is that you need simulations and those are often in themselves compute hungry. So the iteration loop will be slower.
"Our work is so much harder than their work that AI now does" is a refrain of the AI story. In technical terms you concern can be stated as "AI needs to be much more sample-efficient to not be bottlenecked by the speed of doing experiments." People don't find out all the relevant phenomena present there by holy spirit, after all.
BTW, there's also a problem of asking interesting questions that AIs aren't yet good at.
No one has found any principled walls of AI development yet. And empirical results are quite telling. So, I guess, those problems will not stand for long.
I think problem with natural sciences is that it is not so easy to verify solutions to problems - there are always countless competing explanations for the data which is also often noisy - I find AI to lack the "common sense" when working with data from physical measurements .. it somehow has no touch with reality and doesn't have a feeling of the data like a domain scientist
NS is a question for natural science. Q: can we model these bodies of discrete particles with a continuous approximation? A: if you do, you can get aphysical singularities.
"If in other sciences we should arrive at certainty without doubt and truth without error, it behooves us to place the foundations of knowledge in mathematics."
This is a wrong interpretation. Physicists have a shit-ton of models that produce "aphysical singularities", they just work around those to get meaningful answers anyway. This is a whole trope and stereotype. Some of the most successfull and accurate predictions in all of physics come out after you discard a bunch of singularities.
Nobody who actually works in fluid dynamics on any sort of application gives a hoot about the N-S millenium problem. Many do not even know what it is. There is no practical effect of this proof on how we do fluid mechanics.
Whether or not ways exist to work around the singularities, that they exist is surely of note. Before von Neumann formalized QM people were still doing QM, okay fine. But it's wrong to then say von Neumann was doing no physics of note.
"Does there exist a pathological combination of smooth body forces and initial conditions for this set of PDEs, where singularities appear, which by the way is completely impossible to actually create in the real world unless you are a literal God?" is a question of math, not physics. This is a hill I will die on.
If you think about what it actually means to have a time-varying smooth body force defined in all of 3-space, you fairly quickly come to that kind of conclusion.
Even if someone comes up with a construction that does not require any forcing, it is going to be some extremely weird initial conditions that you will never be able to even approximate in reality unless you can move all the individual molecules of a fluid around and set their initial velocities from a far distance.
E.g. the concepts of “keystone species”, or even just the concept of a “species” in general is a lot more gray and continuous than something computationally tractable.
There are lots of startups creating labs that can be managed e2e by agents. That will connect reasoning to the physical world and dramatically speed up the plan, experiment, reflect loop beyond what humans currently do in science R&D.
Maybe. Maybe not. Look at AI drug design - it's not really speeding up the important part - drug trials. There isn't really a coherent plan to use AI for the most complex part of drug discovery at all.
I work in pharma and I've advocated for directly training models that predict drug trial outcomes. I don't think we have the necessary support (data + algorithms) to produce accurate models in this space. I suspect that predicting drug trial outcomes is roughly isomorphic to understanding human biology at a fundamental level (and accepting that our existing models of how drugs work are extremely limited).
There is this infamous xkcd (https://xkcd.com/435/) going like this: sociology is applied psychology -> physchology is applied biology -> biology is applied chemistry -> chemistry is applied physics -> physics is applied math -> math is way up there looking down on other fields
I would argue the main reason AI labs have been focusing on programming is to unlock industrial scale automation, next logical step is to solve math as it's the key to unlock everything else. Once you hold the key for math, everything downstream fields become a matter of compute
Isn't it significant in some sense that this was discovered in China first, and not the US?
In the sense of: we didn't know splitting the atom would change warfare forever, back when Einstein and others published on it. We don't know if there are discoveries of similar impact lying in wait somewhere. If China gets to such a discovery first, we are in Sputnik territory, to say the least.
We knew. That's why Roosevelt started the Manhattan project after scientists wrote to him warning about it, to be sure that the Germans or the Soviets didn't develop it first.
Plenty of things get discovered all over the world and this isn't a particularly earth shattering result. Discovering evidence of a particle predicted for 50 years is interesting but still pretty mundane.
The LHC which is doing much of the particle physics discovery is in Switzerland and France, if you need to be reminded, not the US.
The first strong force theory was published by Hideki Yukawa in Japan in 1935 (which won him the nobel prize) ... so uh, not exactly a sputnik problem.
Sputnik problems (and the nuclear race) are engineering and industrial races, not fundamental physics ones.
> The LHC which is doing much of the particle physics discovery is in Switzerland and France, if you need to be reminded, not the US.
Which is ironically an example of particle collider science happening overseas primarily because the US relinquished its lead on scientific discovery, despite having a similar-scale particle collider practically finished and ready to go [0].
That would be classical 80:20 or more like 90:10 rule, few kms of tunnels ain't close to delivery in any feasible way, just the first baby steps.
The amount of continuous effort (and budget) that goes into just keeping CERN up and delivering is non-trivial. I live nearby and sometimes chat with folks working there (its >15,000 folks although not all do science).
"practically finished and ready to go" is a bit of an overstatement. They made some holes in the ground, and did preliminary research and manufacturing, but it was heavily mismanaged and fantastically over budget, so congress killed it
Ahh... I took "published" on splitting the atom to mean the scientific write up of having split the atom - which was Rutherford - rather than the looser meaning - having just written about the idea in general. My bad.
That is untrue. US science is also funded by NSF grants - or were, before the Orange Idiot and his crew of apes began defunding science they didn't understand.
Much science is done by international (i.e., borderless) cooperation between teams of scientists, as well. Science is becoming, and ultimately should be, nationless.
you are trying to see something that's not quite there in this case. there is shit ton of very good research that happens all over the world. but i also sympathize with what you are trying insinuate and that's also probably happening.
HEP is undeniably an international effort, and it's silly to try and spin this into something akin to the space race. While this is a useful and frankly brilliant potential confirmation of existing theory, that's what it is: confirmation (sort of). Without CERN, without universities across the world including all across Asia none of this would be possible.
Very few endeavors are as truly international as this because the problems are so hard to even explore, and the machines required to do so are the most complex and highly engineered on Earth. Lets not lose that international spirit in science because the blowhard narcissists who run our countries would find that temporarily useful to them.
I think what doesn't get surfaced enough is that the cost of AI tokens is likely to decrease as new computing hardware gets scaled, such as neuromorphic (or later, photonic) computing.
To assume the current cost of Claude Code and the like is largely fixed is not likely to be accurate. The economic incentive to lower the cost of ML compute at both training and inference is very strong -- and likely even stronger once the bubble bursts.
Once the compute cost is much lower, and the systems more advanced as well, then the future of programming changes. The role of the human in this may be more like: we need one senior experienced full-stack, and like 1-3 others, to build and maintain large complex apps reliably.
The strategy for many folks will likely be to wait it out until the next administration when hopefully some amount of sanity returns.
The question buried in much of the detail: there is an indication this doesn't apply for H1B's and similar who work in the national interest or provide economic benefit (presumably substantial). Perhaps this allows an opening for at least some people.
Perhaps the people initiating this -- that is to say, almost universally either immigrants or the descendants of immigrants to the US -- would prefer something like the following version:
With silent lips. "Keep your poor, your tired, your teeming masses, too!
Let me articulate the thing which I believe is on many people's minds:
What is the chance the president will order a nuclear strike on Iran as this war proceeds?
We would hope the odds are vanishingly small, because doing so would be profoundly disadvantageous. But the same was true for initiating this war in the first place. The logic -- such as it is -- of some people in power may lead them to conclude once more that shock and awe can succeed. We've already struck the country with powerful conventional weapons at scale and it has not led to a weakening of Iranian resolve.
All the above said, my personal hope of course is this will never happen. I'm curious what other folks think however.
My real question is not whether it would achieve anything meaningful, but what would be the side effects of such a strike on allies in the region.
I don't have a remotely decent mental model of fallout etc from modern nuclear weapons - my assumptions are they're still toxic enough to be a bloody terrible idea anywhere near someone you like.
I think the main concern would be escalation, e.g. Netanyahu feeling emboldened to use his weapons too. And of course Putin, to try to shock Ukrainian forces and population (good luck).
Alliances might get reshuffled as everyone realizes they need to reassess their nuclear defense and deterrence. It would fundamentally change the nature of modern warfare, not for the better. Let us hope this never happens.
Even if clearly one side is correct without any doubt whatsoever, beyond any question? Such as 2+2=4 -- we should accept a situation where some people insist this is not true? It seems irrational.
I agree entirely. HN tends to be incredibly nitpicky and dystopian. I think it's because so many HNers work in dystopian software-only companies, not doing much in the physical world, away from the algorithms.
Incredible technological innovation is on the horizon. That's why we are not doomed this century. We can make it.
*hits 'reply', knowing there will be nitpicky comments because of course on HN these days, no positive point shall be left standing.
So much advanced equipment is just sitting there in labs, waiting for humans to finally go and make experiments. Which they eventually get round to, sort of, when they can secure funding and when the grad student isn't ill or making mistakes or framing the problem the wrong way.
AI-driven labs can iterate 'good enough' hypotheses way faster than human R&D systems. Automated labs are going to be a major source of discovery.
> Eve independently screened some 1,600 chemicals and modelled how their structure related to their activity to predict which ones were worth testing. King and his group armed the robot with background knowledge and a machine-learning framework for developing hypotheses. Eve then used those elements to design experiments to test these hypotheses and, crucially, performed them itself.
> King plans to use the system — which occupies one-fifth of floor space than Eve does — to model how genes, proteins and small molecules interact in cells. Part of that will involve taking around 10,000 mass-spectrometry measurements each day.
The throughput here is astounding, especially when driven by researchers who really know how to chart a path. I feel every time a critical feedback loop is made both faster and cheaper, it makes everyone participating better. I wonder whether we will see many more "whiz kid" scientific researchers than we have today.
> So much advanced equipment is just sitting there in labs, waiting for humans to finally go and make experiments. Which they eventually get round to, sort of, when they can secure funding and when the grad student isn't ill or making mistakes or framing the problem the wrong way.
That's not really what the article is about though. Short of staffing it with humanoid robots, existing labs and their equipment will continue to be unused.
There are groups that are actively working on automating conventional labs like this. Most of the efforts I know about use non-humanoid mobile robots or even just a six-axis arm on a rail and some lab space reconfiguration
I don't really see why most existing equipment would be usable in this way. When you automate a thing you often have to rethink the entire problem. But more generally, automation is for _repeatable_ things and a lot of research is... not that.
The expensive equipment is usually a small (but crucial!) part of research activity, which involves things like talking to a lot of people, getting permission to do weird or new things, going out into the environment and collecting things in very specific ways, storing and transporting them carefully, observing, etc. Building or modifying existing lab instruments, doing various things with animals that are not co-operative ... and CLEANING. Who does all the cleaning?
Definitely use cases when you have a specific protocol you want to scale, but I'm also not sure how safe I would feel around AI with a license to experiment and access to dangerous reagents, high temperatures, etc. Or, god help us, an oligonucleotide synthesizer. Which is definitely going to happen (if it has not already).
In some cases that would be the same person that does the most advanced innovative and/or creative work.
The idea behind the fully automated system is that fewer hired hands are needed for efforts that are routine enough. But not zero, you still need one person who can do everything at a minimum, if called upon for mission-critical operation.
In the case of the creative work and planning where it is out of the league for AI, these things need to always be done too, but they are not exactly "routine".
Once most of the tedious routine tasks are well-automated though, then the human brain behind the lab can finally relax a bit, with eurekas flowing at the same rate without needing a full 40 or 50 addititonal hours at the bench any more, while even more results are generated than they could do single-handedly too.
Which gives them the time to do the cleaning also, otherwise they would need two humans to serve their only automated system.
I really hope you're right. The challenge with Linux still seems to be practicalities -- like in particular, does Zoom run well on most distributions?
Reports seem to be of system crashes and degraded performance. I imagine there are lots of 'it works for me' stories, but think: for Linux to eat into Windows user market share (which I would greatly support), critical things like Zoom have to work at least as reliably as on Windows. For nontechnical users who would never figure out which incantations to type into the terminal to fix it -- because they have their next meeting in 15 minutes.
I installed PopOS (22) and zoom worked fine right off the bat. So did steam and all my steam games. Heck even my printer worked. (It has since become more temperamental and now only works with one of the 3 print dialogues on my Linux box...)
My game controller worked, my BT headset, the media keys on my keyboard even worked.
Lots of stuff was mildly broken but no more so than it was on Windows. It is just differently broken.
I'm not talking about that. I was replying to a claim that Zoom is less stable on a platform, as if that somehow happened for free and not as a result of a team tracking and fixing bugs on the application side, likely over the course of years.
Even my Starbook, so... literally made for Linux, doesn't do things like going to sleep when I close the lid. It made me switch back to my Mac because despite being able to, I have a life and little time for my main work device to decide to not work randomly.
Linux is never, and I mean never going to be a legitimate alternative to Windows or MacOS on the desktop under the current paradigm. "Switch to X desktop or distro" means less than zero to 99.9% of computer users (probably a few more nines in there too).
"Oh but the Steam Machine!" essentially no one who uses that will actually care what the OS is, it's a shell and a very specific one to do a single task, no-one is buying it as a general purpose machine they can do their taxes on.
Yes, precisely. And then as I anticipated, the "it works for me" stories, even here in this thread. Wish we could get past this steady-state in the Linux ecosystem.
Imagine a Linux distro largely displaced Windows and Mac simply due to usability, security, reliability, and the fact that there's no monstrous corporation pulling the strings. That would be awesome.
Works fine on recent Ubuntu and Fedora, both Wayland and X.
"Fine" and not amazing because occasionally I have screen sharing issues, but that's like once in a blue moon? Could be down to my specific configuration, but it's allegedly more stable than my coworker's zoom on Mac.
I mean... Windows legitimately doesn't work. I work at one of the mag7 and it's a running jokes while using windows that suddenly everyone's microphone quits. We then have to restart. This has been going on years. Our colleagues on Linux don't have such problems.
It's just that we accept windows issues as "that's how computers are". While Linux is expected to work
I haven't used Zoom in years, but Teams in the browser on Linux runs better than Teams natively on Windows. Which is odd, since I understand it is just an electron app on Windows, so it is effectively running in the browser anyway. Still, those of us on Linux have way fewer audio and connectivity issues.
Here's a standard-structure, VC-funded, exit-oriented startup to consider: make video calls reliable. As in, you provide a guarantee and pay the customer if the call didn't work.
Wired headphones could be one part of the solution. They're just far more reliable (if they don't break, which they will). But if the reliability of video calls can be improved so that it's literally as reliable as talking to someone next to you in a quiet room, I bet lots of people would pay for it. There is so much latent frustration about unreliable calls, even with the best setup, even in NASA, in DoD, corporations, zoom and other platforms fail to perform reliably in so many cases.
Funny you mention it; I actually have been thinking of this as a startup/solution for ages (especially since covid). I realized that it's likely a fair bit more difficult (you'd need significant control of both software as well as hardware stacks.)
If you or anyone's seriously interested in pursuing it, feel free to reach out to the email address in my profile page.
> make video calls reliable. As in, you provide a guarantee and pay the customer if the call didn't work.
Microsoft would be ruined, haha. Over the past week, I had about a 30% chance of the call not working and a a 80% chance of the screenshare not working
1. It shows what even this wave of AI can actually do.
2. I wish it were done by different folks, ideally under some kind of public control like NASA research or the NPR model.
3. Keep in mind: natural science is different. It's not always a matter of computation. Computer science folks often struggle with this -- but this virtual world here does not actually exist. Everything is physical, including information. Any natural science PhD or otherwise knows just how complicated nature actually is -- e.g. mention any research topic and try to encapsulate all the relevant phenomena present there. Pure mathematics is different because we define the problem, rarher than explore nature. We are in my view far away from removing humans in natural science R&D. Advancements in AI however can greatly assist us in all natural sciences, which is already beginning to happen.
reply