It's more a case of the fox being in charge of the chicken coop; or bank robbers gaining control of the bank.

The job of the press in a democracy is to hold officials up to scrutiny.

The logic is economic autarky -- the belief that we're always better off having complete political sovereignty over every part of our supply chains than having to depend on imports for anything.

This is likely also behind Trump's trade war with Canada. His demands with Canada aren't about discirimatory tariffs -- those were negotiated away with NAFTA then re-negotiated in his first term. His demands are now infringe on Canadian sovereignty, to give US control of Canada's external trade policy, and even on its internal markets (e.g., the use of French in Canadian internal commerce). Symbolically, that's why he wants to rename Lake Ontario to "Lake America". HIs establishment of US control over the Venezuelan oil industry is his greatest foreign policy achievement -- although it's not clear he understands the legal limits on what he can do with their money.

There's a foreign and military policy angle to this too: Taiwan is absolutely critical to the US economy, so we are absolutely committed to defend them if China invades. Semiconductor autarky would mean it's not our problem.

The problem with autarky is that even if it's a good idea (which economists don't believe), you can't conjure it into existence overnight. It will take many years, possibly decades for the US to replace the Taiwan's semiconductor industry. In effect, the administration is proposing to inflict the very damage on the US economy that a Chinese invasion of Taiwan would inflict.

And, once the US has voluntarily self-inflicted that damage, China would be free to invade Taiwan.

Except time and willpower. However, the McDonald's analogy does hold up -- nobody is ever going to get rich flipping burgers at McDonald's, precisely because it requires so little skill to do that job, making McDonald's workers completely fungible. Since anyone can man a grill, no one is going to command a salary much more than what a high schooler will agree to do it for.

The economic problem with having an AI produce music for you is, you never get the opportunity to learn how to produce it for yourself. At the end of the day, you're paying them $200/month for the privilege of remaining unskilled, and you're competing against everyone else who knows how to enter "write a catchy song for me" into a prompt box. You can't make money that way, because the moment you do, the AI companies (who own all of the talent and all of the production machinery) can simply appropriate your income stream for themselves.

It's the difference playing Guitar Hero and taking guitar lessons -- both let you feel like you're a competent musician, but only the latter will ever get you a career as a musician.

Eh? This is incorrect. Perhaps most fictional murders are premeditated, because that makes for a more interesting mystery to solve, but in real life, the vast majority murders are impulsive, spontaneous acts that happen during arguments, fights, or emotional outbursts (often with the help of alcohol and/or readily available weapons).

You're right, they'll run the numbers and take whichever path maximizes their net profit -- but shuttering (or even heavily modifying) a facility that they've just invested $$$$ into is unlikely to be found to be their most profitable option. More likely they'll try to sell memory chips at a profit of +$.25 per chip in large volumes, because when you've got is a giant hammer, hammering lots of nails is what you can do best.

I wonder if anyone will start selling home systems capable of using cast-off data center RAM? It might be worthwhile if there's lots of that sort of RAM floating around for cheap.

Sure, a company could use an increase in productivity to allow their workers to work fewer hours while getting the same amount accomplished... but capitalism doesn't work that way. Instead, the company will use the increase in productivity to get more work done in the same amount of hours as before... or if they don't, their competitor will, and thereby obtain a competitive advantage over them.

This is especially true for salaried employees, where the cost of the employee to the company is the same regardless of how many hours the employee works, and it is therefore in the company's financial interest to maximize the return on its salary "investment" by getting as much output from the employee as possible in any given pay period.

I've got a co-worker like that; every email I get from him is just him copy-and-pasting text from his pet AI, and when I reply, he forwards my email on to his AI to read and generate the next reply for him.

Which raises the question... exactly what value is my co-worker providing there? Sooner or later management will notice that they are paying six figures of annual salary to what is essentially just a very slow meat-based email relay, and they'll lay him off and just have me communicate with the AI directly instead. There's a fine line between "all in" and "all redundant".

It all depends on one thing: Will your speech-driven AI agent reliably accomplish the tasks that you told it to do? If so, great, we've entered the glorious Star Trek future. OTOH if it screws up, even just 5% of the time, it'll be kicked to the curb, because a "timesaver" that you have to stand over and supervise to make sure it doesn't go wrong is no timesaver at all.

Investors are chasing a company that achieves some kind of AI singularity. Let's set aside the fact that there's no reason to believe there is anything but diminishing marginal returns by making marginal refinements to current frontier models. Let's imagine someone hits the jackpot and gets, not even AGI, but a system that's as far ahead of today's frontier model are ahead of 2020's GPT 2.0 in performance.

Globally AI revenues are 150 billion, against a cumulative burn rate of 450 billion. A model that is a generation ahead of others would almost certainly capture the lion's share of that revenue.

If AGI magically appears as a Sam Altman has promised investors it will, a hundred million is way too low. Add, maybe, another zero to the revenues.

Conservatively, a safer assumption is that frontier models will get marginally better based on refinements in training and reinforcement and the other bits and bobs that go into these systems. The nobody is winning the lion's share of anything, at least overnight. But you have to define "safe". By "safe" I mean unlikely to lose money. But some investors are clearly defining "safe" as "having the greatest chance of owning a piece of the biggest thing ever."

I'm not following this super-closely, but if Anthropic is pursuing adding multi-step model based reasoning to their system, that could be the basis of a generational leap in capability. But if that is an approach that looks like it has a chance of working, then their competitors are no doubt pursuing the same thing. In that case you'd expect the revenue pie to grow as the scope of model utilty increases, but that growth to be split among several competitors. This could credibly result in a revenue stream for some of them that is as big as the entire industry's revenue stream today. But there's going to be hell to pay on the data center impacts end of things.

If the weather is bad enough, people choose not to go out in it. Unless someone has a gun to your head, you always have a choice.

Heh, there's some validity to that. Human drivers are a daily rolling disaster that society tolerates only because there hasn't been any reasonable alternative. If-and-when autonomous vehicles actually become cheap and reliable enough, I suspect that human driving will gradually start to be curtailed -- not necessarily by banning it outright, but by making licensing requirements more restrictive, by making insurance costs more onerous, and/or by "nerfing" the cars such that they simply refuse to perform dangerous maneuvers.

What this country needs is a good two-dollar red-team AI, that will try to create an exploit based on a submitted bug report and use the exploit to compromise a Mac in the QA lab. If it succeeds, the bug report is definitely genuine and should be prioritized.

I agree that has historically been the case, but OTOH Trump is deteriorating so rapidly (both psychologically and in terms of his level of popular support) that after November's bloodbath (and whatever Trump's unhinged response to that will be), the remaining Republicans may well decide it's in their own interest to get him out of office sooner rather than later.