So, $22.50 per day, or $113 per week. How much does one of your people cost, all-in, including benefits? It's unlikely that it's less than $100k per year, and very likely at least double or triple that, if not five times or more. At $200k/year for 50 weeks, that's $4k/week. At the current token price, the AI makes a $200k engineer 3% more productive, the company is breaking even. If it makes them 1.5X or 2X as productive it's a clear and unquestionable win, even with higher token prices.

As for me, think AI makes me about 5X more productive than I would be without it, and that's just considering volume of work. Honestly, I think the quality is a little higher than I'd do myself -- not because the AI writes better code than I do (it definitely does not), but because I'm able to be pickier and do more and larger refactors than I would if I had to do the grunt work myself. Also because I have Claude write documentation that, frankly, I just wouldn't get around to if it were me. The documentation is not nearly as good as if you gave me a dedicated technical writer... but the $2500 I spend per month in tokens would come nowhere close to paying a writer, even if we ignored the coding productivity.

Somewhat, yes (and this most definitely includes the federal government!). On the other hand, small organizations don't have the same opportunities for economies of scale. This is what drives consolidation in most markets; the bigger players can outcompete the small ones because they have efficiency opportunities the small ones just don't, and greater consolidation increases that... up until it gets balanced and then exceeded by bloat, which lets smaller players back in.

That's what happens in competitive markets, anyway. The healthcare market is so heavily regulated that it may not work the same way.

You can only move workers by paying more, either moving them from one job to another or from being unemployed to being employed. But if you're already very close to full employment (which we are; prime-age employment is 83.3%), then in order to get more workers you have to get more people.

Executives, maybe, but "top shareholders" don't make any more than "bottom shareholders", not on a percentage basis. And for all of the shareholders the fact is that they'd be better off selling their shares and buying into a more profitable industry. I'd guess that investors buy health insurance stocks because it's a pretty reliable industry -- it doesn't make much money, but it keeps making money even when the rest of the economy tanks.

You're right, of course, that the ACA establishes a perverse incentive for insurers (similar to the cost-plus funding model NASA used to use), but this doesn't seem to have affected overall costs much. It's most problematic with vertically-integrated companies (like UHC). There is data that shows that they pay more for services that their own subsidiaries deliver -- UHC pays more to doctors in its own groups than to doctors outside of its groups, for example. But overall, the evidence that the ACA had any effect on the trajectory of healthcare prices is weak at best. Which makes sense, because the ACA tinkered with the insurance side of the problem, and that's not where the high costs come from. M4A wouldn't do any better, not unless it used its increased bargaining power to force pay cuts on the provider side.

In a closed system, yes, it means that the current price is too low, so demand is higher than can be filled, which means that the price should be raised, both for labor and for the product of that labor, until the excess demand is destroyed (i.e. buyers can't afford it). However, the system isn't closed. So what it actually means is that the suppliers will move to where the labor is, so they can continue meeting the demand at the current price.

What this means for the US is that US business will at least not be able to grow, and will likely have to shrink.

Which, I have to grant, will fix the labor shortage. Businesses will shrink or close, fewer jobs will be available and labor will no longer be in short supply.

You mean they shouldn't be made here.

Your questions only address one side of the equation, which makes them meaningless overall.

If there is huge excess profit in the system you can expect to see labor shortages causing wages to shoot up astronomically. If there isn't, then what happens instead is the shortage is addressed by businesses scaling back, simply making less and selling less. Ideally, this just results in hobbled growth, but more often it results in businesses trying to do more with less, and failing. That, of course, results in layoffs, which addresses the labor shortage a different way.

And keep in mind that this happens in a sector-by-sector and business-by-business way. So you can't look at Jeff Bezos's wealth and assume that means there is excess profit in lots of industries, just because Amazon is doing well. (And definitely don't look at Elon Musk's wealth and use that to assume there's lots of excess profit; Musk is insanely wealthy because people think his companies will generate incredible profits in the future, not because they're doing it now).

It doesn't really matter what you call workers, if there aren't enough of them, there aren't enough of them.

The rest of your post was just a reasonably-accurate characterization of part of what is causing the potential problem. The fact is that the US economy has always been driven by immigration, and by shutting off the flow of immigrants, the current administration is strangling the economy. It'll take a generation or so to really take full effect, but if we don't reverse course we'll end up with economic stagnation.

This is my read as well. Surveillance has nothing to with what's driving it, though surveillance will almost certainly take advantage of it.

Sure, but that requires positing that this isn't actually Chinese activity at all, but US entities pretending to be Chinese entities pretending to be US people. That's possible, but it seems like it would require some evidence.

The progression in the link isn't what was taught in universities, it was what a programmer learned growing up, and I don't know anyone who learned FORTRAN in elementary and Junior High. BASIC was where it was at (my first code was BASIC to run on a local bank's mainframe -- they were letting my fifth grade "Talented and Gifted" class run stuff -- and then later on a friend's TRS-80; that was all in the late 70s). Then Pascal in high school or college, and so on. Java came later; I'm not sure what the kids who got Java in college learned before college.

FWIW, healthcare insurance shareholders aren't getting rich, because healthcare insurance companies aren't very profitable. The most accurate way to evaluate healthcare insurance as an investment is to look at return on equity[*], and health insurance ROE is 11-13%, significantly below the market ROE of 17.2%. The main driver of high cost in the US is the providers, not the insurers. US hospital and other healthcare facilities in the US have an ROE of over 50% [**]. There is some crossover here; many of the big insurers also own some hospitals. That's the primary reason that United Healthcare has better returns than all of the other big insurers; because UHC has invested heavily in the provider side.

This is why Medicare for all, by itself, wouldn't do anything to lower healthcare costs. It would probably reduce the cost and complexity of billing, which would cut overall cost by a few percentage points. To really reduce costs, it would have to force providers to lower costs.

The reason US healthcare costs are high is because a lot of people make a lot of money... but those people aren't the insurance companies. Doctors make 3-4x what they do in other rich countries. Nurses, hospital administration, device makers... all make a lot more than they do elsewhere. Branded drugs cost 2-3X as much (though generics are often actually cheaper in the US) than elsewhere), which is an area that is obviously ripe for savings... but there's a risk there because those high prices fund a lot of research (pharma is also not terribly profitable; that revenue mostly gets sunk into new drugs).

The insurance companies are the preferred target, though, because the insurance company is the part of the system that gets tasked with saying "no".

[*] In most industries you would look first at profit margins, but that gives a very skewed view of insurance companies in general, because they have enormous revenues (premiums) but also enormous costs (claims). So their profit margins are down in the 2-5% range, alongside grocery stores and other bulk retailers.

[**] The vast majority of hospitals in the US are non-profits, so that 50% figure is based on relatively thin data. However, those few for-profit hospitals compete directly with lots of non-profits, so their price and cost structures have to be comparable. If they charged much higher prices, they wouldn't get any business (the insurers would refuse to include them in their networks) and if they paid much lower wages or bought much cheaper equipment/drugs, etc., they wouldn't be able to hire/buy. So the non-profits have the same structure as the for-profits. The difference is that the non-profits invest the proceeds into their endowments while the for-profits distributed them to shareholders. Predictably, the non-profit investment reserves have grown to be some truly enormous piles of assets.

How does raising salaries fix a labor shortage?

That makes no sense. Why would China care about getting the US population to resist the deployment of domestic surveillance?

No, what China is concerned about is that AI may provide a significant economic and geopolitical advantages to the US. That may or may not actually happen, but it could represent an enormous change in the balance of power. China is trying to enlist the aid of the American population in slowing AI development in the US because it wants time to catch up.

Or maybe you're right, in which case China is confused, because it would actually be beneficial for China if the US were to begin massively and thoroughly surveilling its own population, because it would help to reinforce the idea that such surveillance is normal.

I think you're misreading the intent. This isn't random mockery by individuals, shared with their friends for fun, this is a focused disinformation campaign being targeted through western communication channels. China is concerned about what it might mean geopolitically if AI turns out to be as significant as it might, and the US wins the AI race. This is an effort to slow AI development. A small and opportunistic one, probably not part of the core strategy, but a cheap, easy one that might have some beneficial effect.

I suppose, though I've never worked in any company that evaluated on LoC. I hear they exist, but I've never seen it... and I've probably worked for two dozen different companies, in a wide variety of industries (I was a contractor for a good chunk of my career).