That's not right either. Advertising and reference manuals should be on the internet. Probably some other stuff that I haven't thought of too. But nothing that needs to be private. If you MUST have it remotely available, give it its own phone line, and put protections on that. Dial-up access has its place.

But patching ASAP is also a bad idea. What you really need to do is
AVOID putting dangerous stuff on the web!!
Then you can take time to be sure that the "patch" isn't a real screw-up.

This is probably now wholly absurd. They *have* been sued in the past. I'd guess it was rather low frequency, but that's a guess.

OTOH, my real guess is that it's mainly a political talking point.

Actually there are markets where these might be reasonable. (MIGHT, I'm not real sure of their capabilities.) The ones I'm thinking of are pretty niche, though. Generally hospitals or care homes. (But they probably couldn't pay the costs.)

If US companies lose a market, less money flows into the US. Theoretically that should disadvantage the citizens of the US, but that's based on the assumption that the wealth is shared, even though not equally. A somewhat dubious proposition.

The projections I've seen put China's semiconductor industry a equal to the best in the west sometime after 2030, probably around 2035. This is a spur to convince them to move it closer to 2030, but some of the techniques are pretty sensitive, so I'm not sure it will work.

Your general conclusion is correct, but your analysis is wrong.
Wealth isn't a simple concept. It has no direct relationship to anything physical, but it's got a LOT of indirect relationships.

E.g. part of wealth is being healthy. This depends on physical things, but for most of them there are tradeoffs. Potatoes aren't necessary if you've got wheat or oats. Clean air is valuable, but so is treatment for diseases. And NONE of these things are directly proportional to money.

E.g., part of wealth is having a happy marriage. Lack of poverty sure helps that, but anything beyond that doesn't help much.

The wealth of a country can increase when it stops being "top dog", but this is for sure by no means guaranteed.

Yeah, China is currently getting Nvidia hardware, but it's acting strenuously to wean itself from it. Wherever local stuff is "good enough", then that's what it pushes. And it's pushing development so that local stuff is less limited. I expect this to take them a decade to reach parity, but I've no inside or detailed information. (I'm repeating projections made by someone else.)

Yes, "in five years China will be shipping cheap AI chips all over the world", but they'll only be "good enough", not "top of the line". That'll probably take 10 years.

Well, 10 years still isn't a long time. It will only matter to them if, say, an AGI breakthrough happens before 2030 and still depends on transformer chips. (I've got a theory that the "transformer chips" direction is the wrong direction...well, not *wrong* exactly, but misleading. I think AGI is going to be more dependent on multiple smaller models and close connection to sensors. For that transformers are advantageous, but not really decisive. Check out "SLMs", e.g. https://huggingface.co/blog/jj... )

Sorry, but capitalism has LOS of good points. It's just that it also has more than a few really bad points. Small scale capitalism is extremely good. Monopoly ANYTHING tends to be really bad.

OTOH, some jobs cannot be done without concentrated controlled power. Whether you call the agency that does them a government or a corporation is basically irrelevant, what *is* relevant is the damage they do in the process of doing those jobs.

Small scale capitalism, where there's no significant "barrier to entry" is so far unbeatable for solving problems with few bad effects...IF there's some stronger power that ensures that they minimize the bad effects on others (including those who aren't customers).

Papers published in small journals have been overlooked since before the time of Gauss (who overlooked Lobachevsky).

I suspect that the people who had the AI "solve" the problem were aware of what it had done, but that the PR guys got (or understood) a simplified story. No actual lies were necessarily involved, as there wasn't necessarily any intent to deceive.

OTOH, the statement was substantially false.

When corporations tell self-promoting falsehoods its not necessarily with any intent to deceive. But you still shouldn't trust them when there's not sufficient transparency. And were this a fraud investigation, I think they should be prosecuted as intentional fraud, even though no individual within the system necessarily intended fraud. The system design intended fraud.

Actually Jackson Pollack paintings had (have?) hidden meaning. I don't think anyone knows what that meaning is, but multiple of his (authentic) paintings have been analyzed and there are strong patterns in the way he used colors and the angles at which the lines cross. Presumably that are other patterns that weren't checked for. It's like a coded message that nobody has the key for. But using this you can easily detect fake Pollack paintings.

So, no, you couldn't do an equivalent painting. Just one that looked the same to an untrained eye. It would be easier to do a fake Picasso.

They weren't "utterly false", only "substantially false".
I.e., the problems had been solved. Just not by ChatGPT.

Another reason not to trust EA.