It's not often I say this, but China definitely has the right idea when it comes to executive liability for corporate misdeeds".

Everyone is misunderstanding the intent behind this. The key phrase in TFA is "unlicensed AI output".

Music charts don't measure popularity, they do their damnedest to create it. From the famous RIAA payola scandal (which very much still happens, it's just structured slightly differently to remain legal) to the latest breakout nepo-baby somehow charting in the top ten despite nobody ever having heard of them the day prior - The various recording industry groups exist to promote their members - who are not the actual artists - and couldn't care less about accuracy.

There's nothing human about "the machine" (now an oddly ironic description) fearing AI generated music because it very much is an existential threat to them. I'm neither a fanboy nor hater of AI, but it won't be long before it doesn't suck and these whores can no longer control the market.

Yes - "Thank you sir, may I have another?"

We live in a world where everyone on the planet, on every side of the issue, knows perfectly well 34-time felon and adjudicated sexual assailant Donald J. Trump is at the very least a pedo rapist, a corrupt and incompetent businessman who managed to bankrupt the licenses-to-print-money that are casinos, and was quite possibly Epstein's boss rather than merely a prolific client of the largest human sex trafficking operation known to modern history. And yet somehow, his base is willing to draw a line in the sand over some mythical official release of the Epstein files. The same Epstein files that would 1) completely exonerate him if his story is true, and 2) have already been sufficiently released and/or leaked prior to redactiongate to eliminate any reasonable presumption of innocence.

This is a relatively minor retaliation case. Unless Kegsbreath needs a distraction from his next drunken compromise of opsec, this ruling will be publicly acknowledged with some whining, then privately ignored anywhere it matters. It's a hefty uphill battle to ever "prove" Anthropic doesn't get a particular contract in violation of Lin's ruling.

You realize Netanyahu is Jewish, right?

I can't speak for the GP, but in my case it was a sort of "constructive dismissal" that finally pushed me away from all things FB-related.

I used to enjoy outing scammers on IG. I was even so naive as to think I was in some small way doing the platform a service by sticking a red flag on its bad actors. Whoops, guess what, folks? Scammers drive engagement and that's all these platforms care about. So I started getting progressively longer mutings after every post or comment I'd make, regardless of how tame - It got to the point where I finally rage-quit after getting a three-day ban immediately (as in, the notification arrived exactly as I hit post, so there's not even a chance I was reported that fast) for commenting something like "nice sunset, wish I was there!" to a close friend.

Well, he wants to do it to China, but needs a testing ground first.

I have to admit it's somewhat changing how I write - Not because I use AI to help me write, but rather, my own writing style pretty much meets most of the criteria listed. Case in point, there's a negative parallelism right there and I wasn't even trying for an example. It's an extremely effective rhetorical device, and AI be damned for trying to steal it from us!

Low-frequency word choices, hyphens/dashes/em-dashes (I use hyphens rather than em-dashes, but most people have no idea what the difference is), and Oxford commas? Check. And okay, that last one was intentional.

Accordingly, I now consciously try to sound at least a bit less like AI when writing, mostly through the use of more metaphor and sensory references that have no meaning to an AI.

There is no country on Earth where you're required to own works you cite from. It may be all but impossible to access some content without paying, but once you've seen it - including unlawfully* - you can reference it to your heart's content and there's not a damned thing the author or publisher can do about it.

Google is clearly trying to firewall off their own issues with mass copyright infringement during AI training by passing the buck to the end user. Not happenin', guys! If this rent-seeking scheme actually works in any form, it will just make Gemini everyone's absolute last choice for literature review.

* Yes, there are of course exceptions like classified material and trade secrets, but that's not what Google is talking about here.

35 years ago there were no LLMs / ANNs in cars.

I actually read up on this and thegabz's post about Daimler using NVIDIA and LLMs.

They only do route planning and lidar for "looking around". Why anyone needs an ANN for route planning is beyond me.

The real driving is still done by Continentals ADAS (now AUMOVIO SE).

Interesting is: Continental is in a cooperation with NVIDIA, using its millions of miles visual and other sensorics data to actually train new models on chips made specifically for automotive/automatic driving.

Yes, Mac OS.
It is Sequoia, 15.6.xyz ... not sure about exact version number.

Was a bit complicated to get it installed (because I was working on the other computer, and clicked wrong in the middle).

The working experience is just fine. But in relation to older Mac OS (X)es, Sequoia is not really that fine. The settings dialog, once the most beautiful thing of Mac OS X or macOS, is a pain into the rectum.

At my specific laptop, I have the problem: it does not sleep when closed and plugged into power. There "should be" a fix, according to google Gemini, resetting parameter RAM and some other fix during boot, but I did not try it yet.

Anyway, the Mac runs just fine. And it is lightening fast.

I put the new OS on its own partition, so I can access the old data just fine, but did not risk to damage the old boot partition during install. So I can dual boot, the old stuff (which can not read the new partition because of different file system) and the new OS.

No, my sister lives on the other side of the planet.

No idea what my sister has to do with your bullshit.

It is nto a poor chosen example, as it is the only example I have.
I was on a pilots web site just recently, but got there via Instagramm, so I have not bookmarked it.

Point is: at the moment Russia is avoided, because of the war.

And from Tokyo to London, you would fly over Russia, but not over the pole.

The actual flight is also not over the pole, but passes mid Greenlands.

The discussion was not about poles or not, but about the fact that Russian airspace is either closed or avoided.

Well,

we are not talking about cosmic rays.
We are talking about solar wind. Hence the talk about flying over the poles.

Cosmic ray impacts have no real difference depending on the height above ground.

Sure, there is a big chance that a gamma photon hits something and does not reach the ground, but would hit you 10km above the ground.

However cosmic rays based on ions with close to light speed, they go down until they hit something in the earth core that stops them.

At this point, all models should be tested in air-gapped systems.
a) they are
b) in an air-gapped system: they can not do such exploits - so, you can not find them

Silly idea.

Most current Jew population on the planet are Nazis, you realize that? Right?
Oh, you want to nitpick about the fact that the "National Socialist Party of Germany" does no longer exist?

Nazi is a neologism for: racist, fascist ... get over it.