Except they didn't discover new superconductors. Their AI came up with a novel idea to discover new superconductors, and they didn't even bother to check whether the results it hallucinated checked out.

The distinction between "having an idea" and publishing a paper where you demonstrate that the idea has merit, where you validate it, is one you shouldn't have to really explain to a high school student, much less a grad student publishing their first paper.

This is a magnetic semiconductor (and antiferromagnetic). All non-experimental magnetic semiconductors require cryogenic temperatures. If this pans out (works, cheap-ish to produce), it could mean significantly faster memory, with significantly less energy usage and significantly less waste heat (and thus, even less energy usage).

Obviously it's very far from "panning out", but "room temperature" is not a given here. It's not normal, and it would be a huge deal.

"Whereas traditional electronics are based on control of charge carriers (n- or p-type), practical magnetic semiconductors would also allow control of quantum spin state (up or down). "

https://en.wikipedia.org/wiki/Magnetic_semiconductor

That page lists a bunch of other stuff too.

Wikipedia seems to say it is new, assuming this one has "robust" coexistence of the properties the framing would be important and not just thrown in there to trick people into thinking it was about superconductors:

> To date, GaMnAs remains the only semiconductor material with robust coexistence of ferromagnetism persisting up to rather high Curie temperatures around 100–200 K.

If there is an important combination of material properties that previously was only available at cryogenic temps, a room temp version is significant since you don't need cryogenic cooling to take advantage of it.

That's not really true, I think. (assuming you're talking about "hallucinations" in the sense of the LLM behaviour, not the original meaning of the word)

I was just talking about LLM hallucinations with someone the other day, when doing a brainstorm or discussing ideas, complaining how the confidence with which hallucinations are presented, often led me on wild goose chases that waste my time for an hour or two.

When I realized that, for all the "how is that different from what humans are doing", I have never EVER experienced such kind of confusion with another real human.

And in fact were someone to behave that way, I'd be thoroughly creeped out (you know, the feeling you get when the person across you turns out to be an empty shell of a psychopathic mind, kind of creeps) and do my best to keep them out of my personal circle and avoid having to interact with them ever again.

There is still something in interaction between actual humans, that creates a level of understanding, that LLMs simply cannot (yet?) simulate. There is a certain (deep, unspoken, not language-based) understanding of "this is the other person's goal", instead of "I'm scoring imaginary talking points by talking to them".

And you know the type of person who actually does that, talking bullshit very confidently to "score points", that behaviour is generally considered adversarial and unaccepted between peers, and keeping it up after being discovered to consistently attempt that behaviour, with real humans often results in social exile.

(btw I'm not sure if it's inherent in LLMs that they can't learn this, maybe they can, but my point is that, right now, they are definitely not trained to do this)

In a "three felonies a day" universe?

The appropriate question is whether I want to live in Florida. This is much more a Florida law problem than an AI company problem.

It would apply if you happened to use Office 365 to write your diary.

I don’t live in the US. But this kind of broad law is hard to implement without surveilling all users, and it has multiple side effects.

What happens if I use Claude or ChatGPT to research sensitive social topics? Would that be considered a social network interaction and used against me when I apply for a visa?

Many governments (especially in Latin America) copy what the US does, meaning that similar laws will be pushed sooner or later.

Anthropic/OpenAI is not obligated to report your use, as long as it's not violating some terribly written law (like the Florida law). The government won't know about it, so no, you will not be denied a visa.

> Many governments (especially in Latin America) copy what the US does, meaning that similar laws will be pushed sooner or later.

I honestly don't know, but I suspect most US states don't have such an overreaching law.

Would you use a lawyer knowing they are inclined to turn you into the police if you talk about committing a crime in the future?

The law is the law.

We should be happy about this as for once the AI companies did the right thing.

Does all writing now have to be scanned for thought crime?

To me, journaling your intent in a private journal, whether that's an Apple Journal/Note or a Moleskine in your drawer, feels qualitatively different in some way. But I'm not sure why.

So "I'm going to shoot up the police station" written in your own journal feels somewhat different than "I'm going to shoot up the police station" said to a system that might be able to _interpret_ or _act_ on what was said in some way. Did I just give AI a legal duty, or even a soul I didn't think it had before? I've written up about three or four "what about this, what about that" and deleted them all.

"If you give me six lines written by the hand of the most honest of men, I will find something in them which will hang him."

But that service is a process on MY computer, loaded into RAM, with an abliterated model.

I don't trust 3rd party networks from abusing any data I give them.

It's trained on the speech of online commenters like us; That's where it's getting these. I'm not gonna change just because Timmy the Zoomer has been with his LLM girlfriend for the past three years and occasionally I remind him of her. I'm not going to deliberately code switch into somebody who doesn't open a new tab to Google things as he talks about them. And you can take my bullet point lists from my cold, dead hands.

Not asking anyone to stop using those writing devices, just begging for any work to not be exclusively dramatic devices. And when you read a piece of work like that its generally a child or AI lol.

It was not easy for your local triple A baseball game being streamed to 500 viewers. This approach scales down.

Not to mention human accuracy will likely be way, way worse, no human being will maintain their attention counting ads at five second intervals for even ten minutes, almost two hours is unthinkable.

"Note: archive.org, or web.archive.org, run by the Internet Archive and the most-used web archive on Wikipedia, is uninvolved with and entirely separate from archive.today. Please keep using archive.org"

My ISP blocks both and I’m not even in Spain.

Everything in low Earth orbit is moving at 7600m/s. A ~10g projectile moving that fast has an energy of 580,000 joules. Getting hit by something moving that fast is roughly equivalent to getting hit by a whole ammo case of bullets at the same speed. There is no armor that can protect you; Even the most solid objects behave like fluids at this relative velocity, potentially incurring more damage from the explosion / shockwave splatter than a thinner object that would allow you to overpenetrate.

A cardboard box of ball bearings with a handgrenade in it in LEO being triggered could wipe out nearly everything up there, and depending on the exact altitude it could effectively mine LEO and prevent humanity from using it for a long time, with a selectable duration between months or millennia depending on altitude.

"Space warfare" in LEO is a matter of everybody walking around with a personal nuclear weapon strapped to their back that can be activated on a hair trigger. We may well wipe everything out by accident, if it gets too crowded or because somebody stubbed their toe. Regardless, there is no selectivity, and any actor with space launch capability can blot out our access at will in ways that will outlive them.

Basically our only hope of continued use is things that persuade us to refrain from attempting to incorporate space into a theatre of war. That box of ball bearings should have higher stakes on the escalation ladder than a tactical nuclear strike. And it probably shouldn't be there in the first place.

There are thousands of pieces of debris in LEO large enough to be tracked, and a lot more that's too small to track but big enough to do serious damage. This isn't good but it's not wiping everything out. Adding a few thousand more pieces of debris would not be even remotely as bad as using a tactical nuclear weapon in war would be.

<1000km LEO would take a hundred generations to resume operations.

The collision you note, along with another couple that have occurred, dramatically increased overall trackable debris counts, by thousands, when we only had a few thousand satellites and under a hundred thousand debris items. Secondary collisions were much less of an issue than they are now, and they are now much less of an issue than they will be in five years per current plans.

Also, ball bearings or the military equivalent tungsten BB's, are prefigured for maximum lifespan and maximum distribution, as well as being untrackable from day 1.

The worst effect would be a nuclear weapon in space. The resulting EMP would probably fry more than enough electric and electronic devices to throw back whatever country is hit by it by decades.

Properly aimed at a populated area it would be devastating. And you can detonate bombs far from your enemy's country, then let earth's magnetic field lines transfer the EMP pulse to their territory. Good luck defending against that

That said, it'd take far more than one bomb to take down a country, unless we are talking city states

1 https://en.wikipedia.org/wiki/Starfish_Prime

2 https://en.wikipedia.org/wiki/High-altitude_nuclear_explosio...

Hit the ten biggest metropolitan areas [1] with one nuke each and you'd hit about 88 million people and probably an even bigger share of all industry that isn't farming. The biggest problem is power transformers, there isn't much capacity in manufacturing them, it's already a bottleneck thanks to regular replacement and renewable energy / datacenter buildouts with 24+ months of lead time [2] - now imagine a significant war related impact, the lead time would probably skyrocket to at least a decade.

At least one source claims that the radius from a single device could be much, much larger [3], so with proper design to maximize an EMP (which even Starfish Prime wasn't intended to do), it is conceivable that dropping just two of these things over the East and West Coast respectively would cause immense damage.

A Fallout-style nuclear war, I think, would be easier to recover from than even a small exchange of EMPs.

[1] https://en.wikipedia.org/wiki/List_of_North_American_metropo...

[2] https://berlin.cwiemeevents.com/articles/new-normal-componen...

[3] https://atomicarchive.com/resources/documents/effects/glasst...

Edit: Will the stocks of fertilizer last long enough, before some normalcy sets in again?

Refrigeration is mostly an issue with meat and milk, but eh, humans have managed to do fine for millennia without it.

Fertilizer however, oh boy, that will be a nightmare. We can get by now that we know about stuff like nitrogen fixers or the beneficial effects of livestock like chicken, but it requires a serious effort to rethink mindsets.

¹As folks from PETA and the likes have discovered when freeing them, several times in multiple places. Beginning since at least two decades ago, meanwhile.

Humanity managed to do fine for millennia by existing in a state of frequent malnutrition and a not insignificant number of deaths from starvation. And it was never in a world with the population density we have today.

All of these are more vulnerable today, because of more dependency on electronics, even if they are hardened against shit like that, or in general, but many of them aren't. Which leads to cascades of dominoes falling, taking others with them.

I really wouldn't like to see that tested today at those scales.

We'd bitch and moan and then carry on as usual. Nobody is starting thermonuclear war over fried birds. Maybe someone gets tariffed.

https://en.wikipedia.org/wiki/Outer_Space_Treaty

I don't know how someone like me would even tell if this treaty is taken seriously. For example, going to space is really hard and has historically been somewhat unreliable at the best of times, which may mean that back in 1967 everyone was thinking "we don't even want a 1% chance that someone's failed launch to drop a 99%-usable warhead on a random location where some tiny country or organised crime group can find it".

https://en.wikipedia.org/wiki/Whipple_shield

Whipple shields use spaced plates to disperse the impact of a projectile on the first plate over a larger area on the second plate, and so on and so forth over four or five plates such that the final plate is not penetrated.

My father worked on high velocity impact ballistics and has some very nice multiple exposure results. The light gas guns used to propel the test cubes are just as interesting as the shields themselves.

Armor can protect you against the smallest debris which post the least risk and decay the fastest. Everything else, not so much.

A question for you though - does that insulation vaporizing itself inside a whipple shield generate secondary debris comprised of the whipple shield and condensed aluminum droplets??

What matters is the relative velocity of the projectile towards the target. I practice it will be smaller or larger (if the bullet is incoming). But steering an incoming projectile is way harder.

True. If you really want to mess space up forever you need to launch a rocket full of sand, slingshot around the moon, and gently sprinkle it in an retrograde orbit. That would be fast enough for something as small as a grain of sand to create pinhole-sized punctures in anything put in orbit, and after a few years the sand would be spread out enough to make any launch essentially impossible.

I might have thought about this too much.

A half-hour boost phase through LEO to get to higher orbit is dramatically less risky than spending 87600 hours there over a 10-year orbital lifespan. In the event that it all goes asymptotic, what sort of timescales would be afflicted by collision risks 10^6 or 10^9 or 10^12 greater than what exist now, I am uncertain.

Nope. You just need to make it a high enough probability of your spaceship being damaged that people won't take the risk. For humans that's very low. For anything moderately expensive it's quite low.

There's tons of room in higher orbits.

You have to go through the lower orbits to get there.

Math doesn't math.

The closest solution I've seen is putting 10^6 to 10^9 kg fragmented debris in the 700 to 1,000 km band to create a 5% failure rate at fastest transit, which is economically devastating. But that assumes away grain-grain collisions.

I think when I last ran the math, you'd need to sustain 5 to 10 Starship loads of fragmented debris to that orbit a week to create a proper barrier–near 100% hazard. At which point the solution becomes shooting down those launch vehicles. In a couple years (3 to 10, depending on the solar cycle), the skies would clear to about 1% hazard. That hazard, however, would persist for some time–how long is anybody's guess.

You want an accretion disc around the planet like Saturn?

It's usually much easier to steer an outgoing projectile. :P

The first reason is that outer space is just inconceivably big, even just the space in low Earth orbit (LEO). LEO has very nearly the same surface area as the Earth, which is to say, about 40% more surface area than the ocean. But the ocean averages 3.7 km thick, and LEO is 2000 km thick. So, roughly speaking, LEO is 50 times the volume of the ocean.

That's a lot of space to miss. There's about 15000 active satellites up there, mostly Starlink. Imagine there are 15000 ÷ 50 = 300 divers swimming at random places in the world's oceans, you have a gun that shoots magic bullets that go 7700m/s and never stop until they hit a diver, and you fire one into the ocean at random. How long is it going to take that bullet to hit someone? It traverses, say, 7700 cubic meters per second, but there are 1.32 × 10¹⁸ m³ (1.32 quintillion m³) of ocean it could traverse, and on average it has to traverse 4.4 quadrillion m³ before it hits someone.

It turns out that your poor bullet will bounce around the ocean for about 18000 years, which is the same thing that will happen to one of your ball bearings if it manages to stay in orbit.

The second reason is, perhaps surprisingly, air resistance. The atmosphere thins out exponentially but never quite stops. Most satellites are at a height where the orbits of even macroscopic multi-kilogram objects decay within a few years from air resistance unless you boost them; smaller objects like your ball bearings or the West Ford needles decay more rapidly. Indeed, all of the individual West Ford needles have already deorbited, despite being in MEO at four times the height of typical LEO satellites, though 44 needle clumps remain.

The third reason is the narrow nature of orbital dynamics. The reason you have to go 7600m/s is that you're in free fall, and you have to miss the Earth when you come down, by having moved far enough horizontally that the Earth isn't there any more. If your hand grenade sends the ball bearings flying backward along the orbit and/or downwards fast enough, they'll re-enter the atmosphere immediately. What's less obvious is that the ball bearings that it sends upwards will re-enter the atmosphere after half an orbit. Depending on how forceful your explosion is, only a narrow circle that stay in almost exactly the same orbit may survive.

None of the ball bearings will be lofted to a higher stable orbit. If they couldn't hit the Earth (perhaps because we replaced it with a very, very small black hole) and none of them reached escape velocity, they would all orbit back through the point where the hand grenade went off, once they'd finished an orbit. So (restoring the Earth to our scenario) even after 18000 years, your hand grenade will have only hit satellites at the same height where it went off in the first place, or slightly higher or lower.

The Kessler syndrome is a real, serious problem for space access, but it hinges crucially on the number of satellites up there to participate in the fragmentation cascade.

I'm not that strong on orbital dynamics, but I think the answer is that you're right. The ball bearings launched directly prograde would be in a new stable elliptical orbit with the perigee at the explosion point. That means I was wrong about the debris only affecting the orbital altitude where the explosion happened. But most of the ball bearings would still re-enter the sensible atmosphere within a single orbit.

The Δv required to deorbit from LEO is surprisingly small. In https://www.nasa.gov/wp-content/uploads/2024/06/iss-deorbit-... we find that deorbiting the International Space Station to a controllable place in "a remote area in the ocean" from its current 415 km altitude requires only 57m/s of Δv. The more precisely controlled Space Shuttle re-entry used 90m/s of Δv: https://space.stackexchange.com/questions/33172/calculating-.... It seems crazy that a change in orbital speed of only 1% would be enough, but there it is. There's a lot of explanation in that SE question.

The ISS document also explains that even the enormous ISS will deorbit by itself in "roughly one-to-two years without reboosts", which is why most LEO satellites are in somewhat higher orbits.