And you've done nothing to explain what the use case is. As far as I can tell, the use case is "Someone who wants to use their phone to control the TV instead of the TV remote," which is a tremendous amount of technological overhead for such a negligible benefit.

It's way easier to point your camera at the screen and do an instant sign-in on the TV than it is to get your phone connected to the right Wi-Fi network and cast to the right TV, so the use case would have to be pretty compelling to make up for what a pain in the a** it is when it works, much less when it doesn't.

Major correction here: about something that I have tried to use on many, many occasions, but never used successfully. There's a difference.

You won't read or engage further because you don't actually know any compelling reason to use it. If you did, you would have said what that reason was by now.

Not really, no. If I wanted to use the TV to do all of the networking and playback, I would have just used the TV's app to do it. The number of hotels I've seen where the TV supported Chromecast or AirPlay streaming but did not have a built-in Netflix app are literally zero.

From my perspective, casting is a complete disaster by its very nature. It relies on the display device having full Internet access, which isn't a given. Literally every time I've wanted to do casting, it has been because the TV set's Netflix app wasn't working because of a network problem, and it couldn't get access to the Internet, so I was trying to use the phone's network connection. By shifting the network connectivity back to the TV set, it makes the entire system completely worthless, because the exact situations where it could be useful are the exact situations where it isn't.

That wasn't *all* I said, but it is apparently as far as you read. But let's stay there for now. You apparently disagree with this, whnich means that you think that LLMs are the only kind of AI that there is, and that language models can be trained to do things like design rocket engines.

That's what I just said.

The figures can be entirely correct and still the answer can be "no". Why? Because Android might use the Linux kernel, but it isn't really a Linux distro in any meaningful sense of the word. And Steam Deck and Chromebooks *can* have some reasonable facsimile of a Linux development environment, but I'd expect maybe 0.1% of users to actually turn it on.

So most of those folks are Linux "users" in much the same way that TiVo owners were linux "users", i.e. they are using a device that deep down, at a level that the user is unaware of, runs some small subset of what a Linux distro typically contains, with a bunch of stuff on top that they mostly aren't in control over.

It's like calling Mac users UNIX users. It's technically correct — the best kind of correct — but grossly misleading.

You cannot bypass solving the Navier-Stokes equations with transformers. You will, of course, get some predicted flows with a black box model, and you can, if you choose, claim that prediction accuracy is close enough for 85% of the random samples from your test data, but that will not get you new propulsion physics.

There's a serious danger today that a lot of the science which relies on simulated outcomes is subtly wrong in a way that cannot be rejected outright in peer review, but will take many years to discover later.

Here's where the summary goes wrong:

Artificial Intelligence is in fact many kinds of technologies. People conflate LLMs with the whole thing because its the first kind of AI that an average person with no technical knowledge could use after a fashion.

But nobody is going to design a new rocket engine in ChatGPT. They're going to use some other kind of AI that work on problems on processes that the average person can't even conceive of -- like design optimization where there are potentially hundreds of parameters to tweak. Some of the underlying technology may have similarities -- like "neural nets" , which are just collections of mathematical matrices that encoded likelihoods underneath, not realistic models of biological neural systems. It shouldn't be surprising that a collection of matrices containing parameters describing weighted relations between features should have a wide variety of applications. That's just math; it's just sexier to call it "AI".

Sorry, you seem to think your link invalidates the point. It does not, but you're welcome to make up a strawman and answer it. I would dress up as a batman to keep you honest, but in my culture we don't know what that is. Thanks for playing, though.

... no.

Sometimes, social scientists who are under pressure to publish, anything, no matter what, to increase their publication count, will propose stupid experiments, that don't cost much to do, do not measure any intrinsic behaviour of humanity, and can be modified trivially to generate alternative papers. The trick is to brainstorm and try out a lot of these, until the p-value finally fits.

How much is this problem is down to AI and how much to beautiful tariffs?

Presumably, a super intelligence has ways of identifying the time that transcend the merely intelligent idea you've put forth?

You're suggesting The Matrix is the way to go?

Just out of interest, did DOGE go after the FAA at any time?
We had a Slashdot poster (posting as a/c of course) blaming DEI whenever anything went wrong at Boeing, as you can imagine that was a lot of posts.

They aren't complete idiots, if the solution was that simple they would do it that way.