I ran across this essay, Freddie deBoer offering a bet to Scott Alexander that the AI singularity isn’t coming in the next 3 years, based on a decent decent set of economic measurements:
Massive misallocation of labor in the USA, in office jobs. To paraphrase Ronnie Coleman: 'Everybody wants a big industrial base but ain't nobody want to study metallurgy.'
Currency bid to the moon because the world thinks they should park savings in the US. A giant Switzerland but with trash everywhere and particleboard housing. To do anything about it you'd need to pull a Japan and get companies to stop being so dang profitable, torch the asset values. Tough sell.
Pretty good punt; my personal one is just tons of steel annually. Energy production is good; main issue is countries like Canada and Russia produce a lot of energy to effectively heat the house.
> I used to find it somewhat mind-boggling that allegedly intelligent people think we’re on the eve of some giant technological leap forward.
The majority of intelligent people lack any kind of discernment.
Most managers are pretty dumb and so when I first went into management I had the "smart" idea that we should give the obviously smarter developers more autonomy and we would end up with better results.
Of course we didn't get better results, we got significantly worse results, I almost got fired, and it became obvious to me that most smart people need to be told what to do within narrow confines or they would go completely off the rails.
The devs refused to do obvious stuff like ask the customers, who in this case were in the same building, what problems they were trying to solve.
Great point that this in no way makes our daily lives any better or different. For me it’s a Google replacement and a tutor for home fixits, although YouTube would have been fine. Also a proofreader for typos. Basically a new version of Let Me Google That For You and a better spellcheck.
I recall when some tech CEO predicted that a new development about to be unveiled would change the face of cities. It was the Segway.
The singularity thesis is just as flimsy in the abstract:
- a bunch of people have sacralized IQ to the point where they think it's mana from a video game
- they believe that information is knowledge
therefore
- AI, which has vast information and can do many math problems, has a huge pool of intelligence and it's just a matter of it tapping into the reservoir in a way that spills into the real world.
This isn't even a strawman, every objection to these assumptions you can think of gets papered over with philosophical materialism
If/when this thing goes belly up, some poor latecoming investor is going to find out what kind of belief he was betting on and pitch himself off a building
I didn't call you a retard, though your idea is retarded and doesn't belong here.
You know what they need for total mass surveillance? Fast databases. That, and data fusion: that's it. Fortunately people seem to have forgotten how to write fast databases (Palantir has great salesguys but their tooling is hot garbage). LLMs won't help. They cost too much money to run, and don't do anything useful in querying large amounts of data.
Dipshits with LLM companies are building them because they think they are in control of a technology which will eventually make some kind of AI god which will bootstrap itself into superpowers. This is also retarded, so you're in good company with rich people who believe chowderhead ideas based on a poor model of how the world works.
The stats on US energy usage here are startling but line up with the (flawed) manufacturing index series: https://www.unz.com/bhua/another-way-to-compare-the-worlds-two-largest-economies/
Massive misallocation of labor in the USA, in office jobs. To paraphrase Ronnie Coleman: 'Everybody wants a big industrial base but ain't nobody want to study metallurgy.'
Currency bid to the moon because the world thinks they should park savings in the US. A giant Switzerland but with trash everywhere and particleboard housing. To do anything about it you'd need to pull a Japan and get companies to stop being so dang profitable, torch the asset values. Tough sell.
Pretty good punt; my personal one is just tons of steel annually. Energy production is good; main issue is countries like Canada and Russia produce a lot of energy to effectively heat the house.
Steel tho: https://www.youtube.com/watch?v=MKMG-FdCGtM
> I used to find it somewhat mind-boggling that allegedly intelligent people think we’re on the eve of some giant technological leap forward.
The majority of intelligent people lack any kind of discernment.
Most managers are pretty dumb and so when I first went into management I had the "smart" idea that we should give the obviously smarter developers more autonomy and we would end up with better results.
Of course we didn't get better results, we got significantly worse results, I almost got fired, and it became obvious to me that most smart people need to be told what to do within narrow confines or they would go completely off the rails.
The devs refused to do obvious stuff like ask the customers, who in this case were in the same building, what problems they were trying to solve.
What in the name of God are all the data centers for?
Great point that this in no way makes our daily lives any better or different. For me it’s a Google replacement and a tutor for home fixits, although YouTube would have been fine. Also a proofreader for typos. Basically a new version of Let Me Google That For You and a better spellcheck.
I recall when some tech CEO predicted that a new development about to be unveiled would change the face of cities. It was the Segway.
The singularity thesis is just as flimsy in the abstract:
- a bunch of people have sacralized IQ to the point where they think it's mana from a video game
- they believe that information is knowledge
therefore
- AI, which has vast information and can do many math problems, has a huge pool of intelligence and it's just a matter of it tapping into the reservoir in a way that spills into the real world.
This isn't even a strawman, every objection to these assumptions you can think of gets papered over with philosophical materialism
If/when this thing goes belly up, some poor latecoming investor is going to find out what kind of belief he was betting on and pitch himself off a building
Gulp.
I’m not seeing the connection between efficient data centers in space, and subjugation of the masses. Can you elucidate
Sorry, that's total nonsense. Go away with that bullshit please.
I didn't call you a retard, though your idea is retarded and doesn't belong here.
You know what they need for total mass surveillance? Fast databases. That, and data fusion: that's it. Fortunately people seem to have forgotten how to write fast databases (Palantir has great salesguys but their tooling is hot garbage). LLMs won't help. They cost too much money to run, and don't do anything useful in querying large amounts of data.
Dipshits with LLM companies are building them because they think they are in control of a technology which will eventually make some kind of AI god which will bootstrap itself into superpowers. This is also retarded, so you're in good company with rich people who believe chowderhead ideas based on a poor model of how the world works.