It’s a better measure but not a perfect one. The big problem with the US-China GDP comparison is that the US has much more of a service economy while China has a much more manufacturing based economy.

Manufacturing pollutes much more than services do but services don’t exist without the manufacturing.

That’s why I was saying a better measure would be pollution per GNP. That would cut out services and basically just count manufacturing output. That would make sense because it’s the biggest source of pollution and it’s the source you can do the most about (ie there’s a lot of room to make many parts of the manufacturing chain cleaner).

Nobody is as green as their marketing suggests and China is no exception. China is making huge investments in green tech and there’s still a long way to go.

Because humans just existing produces far less pollution than humans producing a lot of stuff.

It’s trivial to say that a bunch of hunter-gatherers don’t pollute much but we’re not generally willing to relegate people to living in the stone age.

Our economic choices have a much larger impact on pollution than our personal choices do. Ideally we’d have a measure of pollution per consumption. Everyone would have a score that calculates the total pollution created by the entire supply chain that supports their choices. So if a mine in Africa is polluting so a Chinese guy can have a nice air condition, that should be counted for China; and if a factory in China pollutes so that a guy in the US can have a new Iphone, that should be counted for the US.

I’m not aware of any such data set. The closest proxy would be GDP or GNP. That essentially provides a measure of how much pollution the total lifestyle of that population produces.

That’s not really how it works. Some random Chinese peasant (that’s the vast majority of China’s population) doesn’t produce much CO2. You can add or remove millions of them without significantly impacting coal consumption or CO2 production.

Industry pollutes. Some types pollute more than others.

China has been increasing energy usage across the board at a much higher rate than the population has been growing. It’s a nonsense plan because there’s no reason to think that reducing the population would affect that trend.

While there’s a clear trend of China using more coal there’s just as clear a trend of coal making up a smaller and smaller share of China’s power usage over time. Just about every analysis says they’re solidly on track to completely phase out coal by 2025 and nobody predicts they’ll need to shrink their population to do it.

So you’re saying there are just too many Chinese people? How many should there be?

Trains and ships are part of the logistics chain but trucks are definitely part of it. They have a big advantage of not needing train stations or ports, as long as you have a decent road. Some of the larger strip mining operations fill a truck per minute.

China effectively seems to be playing Factorio. They have a solar/wind production rate of X/day and X keeps going up faster and faster.

They’ll sell those panels and turbines to whoever will take them. They’re cheap but the sheer volume means that you need a huge economy to take any significant share of that inventory. With the US effectively out of the picture the biggest remaining economy is China. On top of that the EU does have some tariffs on Chinese renewables and that skews the deployments even more towards China.

Unreliable may have been a poor choice of words.
You can’t move coal around with pipes or wires. Someone needs to drive trucks full of coal to a power plant.

The pollution from coal tends to have a lot of externalities that drag on the economy. Lost work days, faster equipment degradation, etc.

They use coal but they have practical reasons to want to reduce reliance on coal.

That’s a very emphatic restatement of your initial claim.

I can’t help but notice that, for all the fancy formatting, that wall of text doesn’t contain a single line which actually defines the difference between “learning” and “statistical optimization”. It just repeats the claim that they are different without supporting that claim in any way.

Nothing in there, precludes the alternative hypothesis; that human learning is entirely (or almost entirely) an emergent property of “statistical optimization”. Without some definition of what the difference would be we can’t even theorize a test

GDP is total production net of total consumption. It would be cool to compare it to those factors independently but don’t know of anyone who reports that data.

I’m not looking to bestow sainthood upon any country. Just looking for the most accurate metric.

95% of the world’s new coal construction (2023)

China had the largest new coal construction in 2023 but it was far below 95%. I didn’t do all the math but it drops below 50% when you compare it to just the growth of the next three biggest coal producers.

They build most of our solar but we’ve effectively banned it now. They’re not only growing capacity to produce renewables, they’re taking the outputs that were planned for sale here and installing them locally.

Yes. And go check the percentage of coal use over time. Coal is going up. Renewables are going up much faster.

You should be pretty happy with China then. They have a replacement rate just over one. That’s lower than the US or Europe.

This has been going on for years and will continue.

China really really really needs a robust and diverse energy infrastructure. Industry needs huge amounts of energy. AI needs huge amounts of energy. The military needs huge amounts of energy.

Coal is unreliable and dirty. Oil can be blocked at the Straight of Malacca and a few pipelines.

China is also the world’s factory. They own the entire logistics chain for producing renewable generators; from raw materials to final assembly. They have all the infrastructure to not only build solar panels and wind turbines at scale, they’ve scaled up building the machines that build them.

Pollution per GDP is a better measure. https://ourworldindata.org/grapher/co2-intensity Pollution per GNP would be even better but I can’t find it.

Individuals don’t pollution much, it’s mostly industry. Really poor countries often don’t pollution much because they can’t afford to. Sometimes they pollute prodigiously because the only thing they can afford to do is destructive resource extraction. Rich countries can often outsource their pollution to poorer countries.

China has been making mind boggling investments in renewables. They have been expanding all their energy sources but their renewables have the lions share of the growth.

They’ve been building roads and all kinds of infrastructure. That’s what the BRI is all about, even if they’re being a bit quieter about saying the phrase. They like to build their long haul roads on elevated columns; not only because it’s less disruptive to wildlife but because it lets them use giant road laying robots to place prefab highway segments.

They dropped the one-child policy a while back but they’re having some trouble getting people to have more babies. That said, there’s some research that suggests that rural populations around the world are severely undercounted, so they may have a bunch more subsistence farmers than they, or anyone else, realizes.

So you’re confident that human learning involves “understanding” which is distinct from “statistical optimization”. Is this something you feel in your soul or can you define the difference?

Human learning requires understanding, which AI is not capable of.

How could anyone know this?

Is there some test of understanding that humans can pass and AIs can’t? And if there are humans who can’t pass it, do we consider then unintelligent?

We don’t even need to set the bar that high. Is there some definition of “understanding” that humans meet and AIs don’t?

You’re correct that a collection of deterministic elements will produce a deterministic result.

LLMs produce a probability distribution of next tokens and then randomly select one of them. That’s where the non-determinism enters the system. Even if you set the temperature to 0 you’re going to get some randomness. The GPU can round two different real numbers to the same floating point representation. When that happens, it’s a hardware-level coin toss on which token gets selected.

You can test this empirically. Set the temperature to 0 and ask it, “give me a random number”. You’ll rarely get the same number twice in a row, no matter how similar you try to make the starting conditions.

You may be correct but we don’t really know how humans learn.

There’s a ton of research on it and a lot of theories but no clear answers.
There’s general agreement that the brain is a bunch of neurons; there are no convincing ideas on how consciousness arises from that mass of neurons.
The brain also has a bunch of chemicals that affect neural processing; there are no convincing ideas on how that gets you consciousness either.

We modeled perceptrons after neurons and we’ve been working to make them more like neurons. They don’t have any obvious capabilities that perceptrons don’t have.

That’s the big problem with any claim that “AI doesn’t do X like a person”; since we don’t know how people do it we can neither verify nor refute that claim.

There’s more to AI than just being non-deterministic. Anything that’s too deterministic definitely isn’t an intelligence though; natural or artificial. Video compression algorithms are definitely very far removed from AI.

That’s a reasonable critique.

The point is that it’s trivial to come up with new words. Put that same prompt into a bunch of different LLMs and you’ll get a bunch of different words. Some of them may exist somewhere that don’t exist. There are simple rules for combining words that are so simple that children play them as games.

The LLM doesn’t actually even recognize “words” it recognizes tokens which are typically parts of words. It usually avoids random combinations of those but you can easily get it to do so, if you want.

That’s possible but it’s not what the authors found.

They spend a fair amount of the conclusion emphasizing how exploratory and ambiguous their findings are. The researchers themselves are very careful to point out that this is not a smoking gun.