I'll go ahead and guess that this is Meta training/testing their AI in production for moderating accounts/content on the Meta platform.

Maybe instead of "Terminator" he should mix up something based on "Detroit: Become Human". Might require some care to keep it from turning into "I, Robot", but I know I'd be interested in seeing what he might come up with.

Given the topic, this struck me as an amusing qualifier. Based on the paper's conclusion, said paper should sail through the peer review process with flying colors, given it's probably citing EVERYONE! :D

I don't entirely hate the idea. VW gets a benefit from simplifying hardware variation, just unlocks via software.

Summary is unclear. The article also notes that Volkswagen did not clarify whether the upgrade is tied to your account or to the car.
If it's to the car, then....*shrug*. If it's to the user...I guess it depends one whether that upgrade is applicable to future cars (seems unlikely).
What I would fear would be treating it like certain console games where you could sell the game disc but certain features (included DLC, content, or w/e) would be tied to a one-time code that came with the game and the purchaser would have to buy a replacement code from the publisher. They can fuck right off with that bullshit.

This is true when the parameters are ambiguous. Assuming the command issued was something along the lines of "move foo bar", an incredibly simple trick is to append a backslash change it to "move foo bar\". The command would return an error about not finding the specified path.
Also, don't force overwrites with "/y" when you aren't planning to overwrite data. Of course, the "AI" could just as easily have been running it interactively and just bulled its way through the unexpected results.

This is a fair and proportionate response to such a ridiculous policy. The idea of a nation taxing a foreign company not operating within their borders for transactions with (or involving) their citizens is preposterous, and it's curious how they even conceived of such a hare-brained cash-grab!

That was, naturally, sarcasm. Uncle Sam *always* wants a piece of the pie. For instance, foreign streamers that monetize will (generally?) be taxed on earnings resulting from viewers in the US. I find this specific example interesting, but, more broadly, "foreign corporations" profiting off US citizens/companies will be required by the US to pay taxes on "Effectively Connected Income" (and possibly non-ECI, as well).
What's good for the goose...

In their defense, the comment to which they were replying could have benefited greatly from a second glance at the punctuation. The meaning was still reasonably clear, I agree; however, every comma in that post would have worked better as a period.

I'm glad I actually skimmed the article before posting, as I drew an inference from the summary that just wasn't true.
A neat idea, if they can make it practical, but I didn't see anywhere they mentioned the size of the apparatus to generate this result. How long would it take, I wonder, for the cost of building it to pencil out?

Didn't realize TCL had premium TVs. Only ever seen junk with their brand on it.

So dark....so delicious! :D

I was going to make a joke(?) about shortening that timeframe with another war, but I like yours better, haha!

I believe "client-side", in this context, is referring to the game application itself as a client of the online server hosting the session. In which case, there would be no concerns beyond normal application security.

Saw the title and my first thought was: "To better detect non-HP ink cartridges so it can refuse to print?"
To protect consumers from poor print quality, of course.

They should have asked the AI to help them design these computers for gaming. :-P

On a somewhat related note, the current state of AMD64 -> ARM64 in Windows 11 seems pretty decent, at least for some basic use cases. I had a client who wanted to run a piece of Windows-only software on their new MacBook M3. Originally, I tried the app a Windows 7 VM using x86 emulation, and it was trash. Not unexpected, but had hoped it wouldn't be quite that bad. Loaded Win11 ARM in a VM, instead, and it ran the x86 app just fine.

Ran Memtest86 on it for 3 days, never reported any errors. I presume that the instability only manifested under loads that were pushing the power usage above the limit (that is now) recommended by Intel. Memtest86 may not have pushed it across that threshold. From what I've read online, very math/arithmetic-heavy operations were more likely to be affected.
Didn't particularly test different archives / compression algorithms. The system could definitely extract some other various archives without error, including other LZMA2 archives. Checking the archives I tried, now, I see that the few similar archives I checked only had a 16MB dictionary, not a 64MB as in the NVIDIA package.