Surprisingly, this isn't necessarily true.

To take an intuition: there are lots of problems that are easier to pose and to check than they are to solve, in mathematics.

Disagree lol.

We could move the entire population of San Francisco into full time care for very little compared to what we're spending on AI.

They'd have to consent.

They're discarding the cost per mile to a driver. If you add that back (and especially if you add externalities back) driving is more expensive.

You're missing a lot of externalities here.

Also, the cost to a driver of the car is around $0.70 not including cost to the government and cost of externalities.

Sounds like someone was cruel to you at one point. The parent likely means the opposite of what you're suggesting.

Most people consider mathematical intuition something that only develops from deep study. What the parent is referring to is being able to anticipate how a system will behave before you actually do the calculation.

Most people who use mathematics in an industry job rely on their intuition rather than rigorous proofs. And most industry mathematicians will acknowledge that they're not as rigorous, nor as academically gifted, as their friends who stayed in academia.

Just like a programmer with 20 years of experience can look at a bug and anticipate the cause.

You understand that the US has alienated almost all of its trading partners and is aggressively attempting to kill free global trade, right?

Whereas China is developing a powerful network of rapidly developing trading partners like Africa and the Middle East?

Yep, exactly. US is in the active process of shooting itself in the foot. My point is that China is very dependent on resources imports, so there's no way for them (and any other country) to be self-sufficient or they will stagnate.

Uh, in the housing that they build? The amount of housing we can build with cheap immigrant labor is so much more than what we can do without them. Housing gets more expensive when immigrants are pushed out, because the cost to build rises more than the reduced demand reduces prices.

Partly true, though to a first approximation, the reason we don't build more houses isn't an issue of construction labor costs. Many places with jobs don't allow building homes.

(That's not inherently a universal problem, especially if you're talking about the entire labor market. But it's a problem in many places.)

Schmidhuber rarely ever executed the idea correctly, which makes his claims particularly obnoxious.

No, pangram is fairly decent, unlike the others.

Maybe, but when I used it recently on my annoyingly AI-written something it said 100% human.

They do deliberately err on the side of not accusing people of using AI when unsure. I think that's a good decision.

Do you know it was actually AI? I've seen humans imitating AI "it's not X — it's Y" to try to fool it that it correctly guessed was still human. Humans did use those patterns before AI so they're not enough on their own to flag.

Yeah, because I had generated it, and I wanted to see if I could get it to sound more human with pangram’s crit.

I don’t know if I’m just a lot more sensitive, but I find it really hard to write up even basic communication with Claude. Super overwrought.

Nope. Sorry. Not how this works.

How does it work then?

I'm going to cherry pick one example where newer models are noticeably improving at least in my experience.

What is a noticeable improvement with something that struggles to read a message longer than 200 characters without missing information in the middle, may be a 0.000000001% improvement with a model that... almost never misses info in the first place.