There are some physics effects that can rather neatly be modeled as 'negative temperature'. LASER emission is one: simplifying a lot, you can use negative temperatures in formulas describing usual radiation transfer and you get a working description of how laser emission works. Temperature is a bit of a tricky measure to define: you don't have to move to the realm of exotic physics of find conditions where any definition of temperature starts to behave in nonintuitive ways. For example, we do not tend to think of temperature as a property that depends on the direction: air temperature measured northwards is no different from temperature measured eastwards. But that's not strictly speaking true: the particles forming the atmosphere do interact electromagnetic fields, meaning they can more more freely in some directions, less so in others. The difference is insignificant in our atmosphere, but once you have a lot of charged particles around, 'temperature' can be 100 Kelvin in one direction, 100,000,000 K in another.

No, probably not. Water in itself is a *very* common molecule in space and it can occur in wildly unhospitable conditions. And, of course, even if the physical conditions are somewhat sane, water is hardly the only thing you need for life. There is at least traces of water pretty much everywhere in there solar system, including the sun itself (http://solar-center.stanford.edu/sunwater.html). As far as we know, there's life only on Earth in this system.

If the aim of the 'AI' in question is to create human-like output, not being able to separate fact from fiction counts as a success!

"in March GPT-4 was able to correctly identify that the number 17077 is a prime number 97.6% of the times it was asked. But just three months later, its accuracy plummeted to a lowly 2.4%. " This is 'getting worse' only if one assumes being factually correct is the main development target for ChatGPT. Which, as far as I know, is not the case. If the main objective of ChatGPT is communicating like a human, NOT being able to identify prime numbers is an improvement :)

Fission is a type of radioactive decay: spontaneous fission occurs quite naturally. Induced fission, on the other hand, is mostly limited to man-made nuclear reactors, although we do know at least one case of a natural nuclear reactor: an unusual combination of very high radioactive isotope density and a moderating groundwater flow created conditions sustaining an induced fission reaction without any human involvement: https://en.wikipedia.org/wiki/...

Trains do not, at least here, check if all passengers are onboard. The train leaves when the train leaves and if you or your stuff are in the wrong place, tough. Air travel operates on a different set of rules: they keep track of each person supposed to be on a flight (and, hopefully, their luggage). If someone disappears, the airplane staff will know, and it may cause delays as the plane has to wait into the last second rather than leaving as soon as everyone is onboard. There's also the security aspect. A missing passenger is a potential security problem. If their luggage is still in the plane, it has to be removed, causing further delay (takes about 20 minutes in my experience). Even if they had only carryon luggage, they were in the cabin and now they are not. Did they leave something deadly on the plane? You'd like to know.

Pretty much. Creating a programming language is not an effort that takes hundreds or thousands of people: it does not cost millions to do. It is a quite natural next step from really mastering a few programming languages. The effort needed to make your own language 'perfect so task X' or 'implements novel feature Y' is modest in comparison with other major engineering tasks. So people/organizations do it, and some of the languages take off.

True ... but that's not what the research actually says. Phys.org is pretty much the Daily Fail of science reporting: they get things wrong with a monotony that makes you wonder if it's on purpose.

From the summary of the research paper: "observations of the distant cosmos should be time dilated and appear to run slower than events in the local universe.". Appear is the key word here. Time did not run any slower then than now. It's just that when we observe the light that has travelled across the universe for a loooong time, we see the effect of the cosmological redshift. It has nothing to do with the conditions when the light was emitted.

It's not a pure coincidence. But maybe not in the way you think. it's always worth reversing the claimed direction of the cause and effect to see if the opposite sound sensibly. News tend to present finding the wrong way around as that makes them sound more surprising and click-baity: 'People with liver issues more likely to be heavy drinkers, study says!'

In this case, it may well be that the lab was there because of the viruses. Where would you build a research laboratory working on known potential zoonotic viruses?

Perhaps in an area where your research subject is abundant and where there is a long-recognized potential for problems?

Anyone with 100k can in practice trivially turn it into 1M. With 100k, you should be able to find a bank that offers an interest higher than the expenses of the account. Then you wait. It may take a while though. Sure, there are some edge cases where this would fail (like negative interest conditions), but in the long run they are rare enough not to matter.

To test this, I asked one AI "How would you waste 900.000 dollars". The core of the answer was

"I can suggest using the money to invest in a good cause, such as supporting charitable organizations that address important issues like poverty, education, health care, or environmental protection. Alternatively, one could use the money to travel the world, learn a new skill or hobby, or start a business ventures. "

Those are ways of spending money, not wasting it, dear AI!

aaaand that is why the thegarbz said "clinical depression".

None of the quotes say anything about producing energy for sale. Forbes gets closest to that claim, but that's only because it's the least definite one. "On a commercial scale" can mean almost anything. Does it mean 'lots of peak power'? Does it mean 'predictably'? Does it mean 'with little downtime'? Who knows. ITER, as an experiment -- if successful -- could well produce lots of energy and even sell it, but it would certainly be only occasionally available. "We can produce city-level power output but we'll be available only 1% of the time" is not commercial energy production even if your output while running is very high.

Just in case someone actually thinks we have anyone using "rakes to clear out the leaf litter" in forests here in Finland .... we don't. Never heard of it. It's one more thing Trump lied about just because no-one can check these things, right? https://www.theguardian.com/us...

A 5-fold increase over 50 years averages to a 38% increase every 10 years. If we are going to have only a 50% increase in the next 27 years, the rate of increase is less than half of that of the past. It's still increasing, but much slower than before. Which is, kinda, good news.