My indirect point wasn't that they all should be shut down, but rather this smells like China bashing or boogymanning (boogytizification?).

I haven't seen how this alleged breach is related to the origin of TikTok, but maybe I missed something? It's something China can do to the other sites also.

Betcha 100 bucks Xi also grabbed Facebook and Xhitter user info. Most commercial sites are not fortified against deep-pocketed state-sponsored hackers, only typical private hackers.

Bob, you lived a DRAM good life!

Dammit, I told them to STFU about it; I need free laundry, with the expensive tuition and all. Next ya know, they'll spill the free Top Ramen and condom dispenser bugs. Those are show-stoppers for me.

What "fundamental" AI research requires $30 billion? I'd like to see the proposal first, otherwise the gov't will waste it on silly crap if it's too open-ended.

In other words, I don't trust gov't with a 30b blank check, and I'm a liberal.

No, I'm saying it was a lucky coincidence due to the way the domain changed, or didn't change. We can't expect lucky coincidences to happen often.

Where the gov't often does help is with forming standards to reduce vendor repetition and incompatibility. It may take R&D to test such standards, with the help of industry.

But hiding such research is only allowed for secret projects. Commercial AI wouldn't be able to use it. So who are you expecting it to boost? I don't believe Space X got special access to secret info, but don't quote me.

The Cloud business is kind of like the nuclear power business. Even though the average risk is probably lower than the alternatives, mistakes make the news and quickly damage your reputation. All the big cloud vendors will probably suffer at least one embarrassing outage or breach.

[disbanded] research team that would prepare for the advent of supersmart artificial intelligence capable of outwitting and overpowering its creators.

They figured they were SOL, so instead bulk-ordered tombstones for each team-member that say:

Here lies an actual human. Sorry the bots won. We couldn't put GenieGPT back in the bottle, but are truly sorry for letting it out.

The COBOL team did something like that. Turns out COBOL became a zombie and runs the back-end of civilization.

Actually COBOL survived because they ended up just cloning most of Grace Hopper's and others' beta languages when they ran out of time, since those betas were working in practice, and thus the concepts were road-tested.

I'd still have to buy 2 Windows licenses. I suppose it's worth it.

fiber buildout of the 90's. The amount of fiber they laid turned out to be massive overkill because of gains in throughput per fiber

But web traffic eventually grew to use up most that excess. It just happened after the investors got screwed due to bankruptcies (and picked up by other co's during fire-sales).

Same with the railroad boom of the late 1800's: overbuilt, bubble popped, investors got bleeped, but the rails eventually ended up being used as the population of the US gradually increased.

Thus, AI will likely follow a similar pattern. Right now big-tech is fighting over AI market-share, subsidizing their projects. Such corporate subsidies can't last forever, so it's a financial game of chicken. When investors get a clue and most AI projects are cancelled or sold to the winners, then the excess capacity may gradually get used for actual profitable products.

One problem though is that the motherboards may be obsolete by then. But at least the rack & wiring is already built.

(Some of my own stocks ended up partly tied up in this, so I have to ride along...)

My Wintel PC's consistently last for 6 or 7 years, after 4 generations of Windows. In same cases they are perhaps repairable, but I usually figure it's close to enough to upgrade time to not bother. Windows gets cruftier over time, probably from ether malware or left-over parts from removed or upgraded software and/or malware. Only one replacement was due to a storage-drive failure.

MS has a reverse incentive to make Windows long-term-friendly. If it breaks, they get a new sale.

Perhaps I should get 2 PC's: one where I install experiments and questionable software, and the other for rank-and-file tasks: document editing and browsing. That way dodgy-ware doesn't break common-ware. But drives tend to fail after 7 years anyhow.

I would hope they only do it when the wind is blowing out to sea.

But why don't they rent a small remote island(s) to test? I realize hauling equipment around the globe ain't cheap, but fewer can and will complain, since there's no people around.

DNA is expensive to maintain, from a biological stand point. Why would nature select for simple pairs chromosomes but fill them with 90%+ junk. Why not "spend" your DNA budget on tetra+ chromosome groups and less junk?

Because such clean-up requires a mechanism, which doesn't exist until it gets evolved. Carrying along some broken-and-disabled junk in long-term program storage doesn't cost enough to be a strong selection pressure, so under just mutate-and-select it can hang around for a long time.

Also, if it is the remnant of a functional subsystem, until it drifts too far it might again come in useful, randomly get turned on again, and provide an advantage. After it drifts into doing something somewhat different, it might get turned on again, prove adventitious, and become the basis of a new system.

Lifeforms don't have to be built to superb standards. They only have to be good enough to work, and to work well enough to not get squeezed out by something that works better.

So it's entirely plausible that a lot of DNA junk might be present.

Which is not to say that there ISN'T one or more DNA-cleanup mechanisms currently in place in some branches of life. (Probably evolved from old retroviruses, like several known mechanisms for DNA alteration.)

But it's a hard problem. (How does the mechanism know not to take out something necessary? If there's something it avoids chopping out, what's to keep junk DNA from masquerading as good DNA to hide from it? Junk DNA subsequences are still subject to evolution and hiding from a DNA cleanup mechanism is advantageous to the DNA.)

I remember hearing scientists saying most of DNA was junk left over from evolutionary processes.

I recall, back in the sixties, a co-worker speculating that, considering the genome as a program and presuming a creator (or creation engineering team), the non-coding DNA might be the program's comments.

And wondering whether, if so, they should be considered "holy writ".