api

I'd say rationalism and technofeudalism/neoreaction are adjacent more than directly linked. They seemed to develop in parallel around the same time and involved regulars on some of the same boards.

... and they're at least releasing the results.

This is part of why I can't feel bad for them. The training data is mostly pirated. Whining about Chinese labs training off American frontier models is "waaah you pirated my pirated stuff!"

The tech itself is amazing and fascinating and cool, but the industry is a mass piracy operation.

The biggest exception is cloud. Big cloud carries an insane markup (bandwidth is like 10000X!) and everyone runs on it.

The strategy there is false openness where deployment complexity is the real proprietary moat. Sure Linux, Docker, Kubernetes, Postgres, and all the other standard tools in the box are open source and free, but they're also arcane and complex to run and hard to make fault tolerant. So you're lured in by "open" and then locked in via a kind of "death by a thousand cuts" complexity moat.

(Personally I hold the view that complexity and arcane-ness beyond a certain point is indistinguishable from closed in practice. Open source that's really complex and hard to run is not open in any meaningful sense.)

AI may not admit that kind of moat though, because AI is very good at slicing through that kind of thing. You can prompt a model to make itself compatible with another model or to change code to make it compatible. There's no moat because the moat bridges itself.

Doesn’t matter for training or most AI use cases. We are talking tens of ms latency here.

I’m gonna guess availability of employees may matter more.

Are you sure that works against the mythology of America?

Questioning authority and mythology is very American.

The problem is that this kind of low-tier media slop work is what a lot of artists, writers, etc. do as "potboiler" work. It's what pays the bills.

Historically art of any kind is a U-shaped market: there is low-end work and high-end work. Nothing in between.

So I do understand some of the AI hate among that population. It's chopping the bottom tier work off. Either you're a top-tier massively successful artist or there is $0 to be made anywhere doing anything.

Long term I think it will do that to all white collar work. There will be no entry level jobs. Period. None. Zero. You're either very experienced or there is no work.

This is a huge problem, and one we will have to address.

Lots of devs know how to code in it, and the AI models have more training in JavaScript than any other language.

It's like asking "why does everything run on Windows?" Because everything runs on Windows.

Anyone else think the AI environmental backlash is astroturfed?

I keep looking at the numbers. The power use numbers are not that problematic. Ordering a burrito on DoorDash uses more power than a few days of heavy AI use. The water argument applies to some locations, and is mostly a local governance problem... if the data centers are using too much water, it means they are not being charged enough for that water. Charge them more and they'll push toward closed loop cooling.

Yet the visceral pile-on here is so extreme, it feels fake.

One thing I've learned after 40 years on this planet is: propaganda works, and much of what a large fraction of people believe across the entire political spectrum (left, right, anything else) is there because someone paid to put it there. It's depressing but it's true, and it makes sense. Propaganda is an asymmetrical attack on human cognition and discourse, and in information security the attacker always has an easier job. Crafting viral bullshit is orders of magnitude easier than fact checking. On top of this, humans are busy and don't have time to fact check and logic check everything they read. As a result, much of what we believe is "sponsored content."

People get mad when you talk about this because everyone wants to believe they're too smart to fall for propaganda.

In any case, the US AI labs deserve to lose for their stupid "safety" regulatory capture monopolization push, which ended up blowing their own feet off and handing the lead to China.

The dumb efforts by the US AI industry to use fear mongering for regulatory capture will hand dominance to China and others.

In a few years there will be Mythos level open weight models hosted by the lowest bidder anyway.

This is a weak argument. Genetic engineering and artificial repair of major genetic disorders and artificial ways to boost genetic diversity are all way way easier than building a starship.

In Gibson’s Neuromancer AIs burned onto chips were literally discussed and called “ROM constructs.” In that case some of them were images of human minds but with no persistent state or memory.

So basically a static model version of consciousness uploading.

That project was cancelled. This is mostly about workplace surveillance.

Most of the replies are large pop subReddit level junk.

“AI” in title gets clicks. So “AI” must be in title.

Every time “relentless push for efficiency destroying X” comes up I should repost this:

https://sohl-dickstein.github.io/2022/11/06/strong-Goodhart....

This is a well understood phenomenon.

I had a thought a while back: sell large local models burned onto fused compute / ROM chips. Like cartridges for old game consoles. Slot (or probably plug into USB-C) and go.

It’s an ASIC with the model wired into it so it’s very low power and fast.

I’d buy these. Say $100 for a frontier class model. Maybe more.

Mobile maintains good battery life mostly by being idle almost all the time, and P2P can’t work that way. Even P2P with central server assist is hard on mobile.

It could if we’d never broken the full end to end principle and put stateful firewalls and NAT in front of everything, but that ship has sailed. All net connections require keepalive or a firewall will close them. This is even true these days in the cloud for the most part unless your machines are “naked” VMs directly on the globally routed Internet.

I need to qualify my "I don't care if there are quadrillionaires" comment.

In theory, all other things aside, it is true. I mean, look up at the stars. For all I know there are functionally immortal super-minds out there who have not known pain for millions of years. Their existence is not a problem for me.

But "we live in a society," and unfortunately other peoples' extreme wealth can cause problems for me.

One is by bidding up assets. The super-rich park their money. Parked money goes into things like PE. PE can't find enough stuff to do with it, so they do shit like buy up housing and hold it off the market to profit from real estate appreciation. That makes it harder for me to find a good place to live. That's just one example.

I actually coined a term for the above: financial pollution. Financial pollution is any time someone else's money being invested does harm rather than helps, usually by driving up asset prices, massively distorting a market, or being invested in the creation of rent-seeking schemes. A classic example of financial pollution would be the real estate market in cities that allow a lot of overseas investors to park money in real estate.

Financial pollution is the opposite of productive investment that does things like creates jobs and builds things.

Another is that extreme wealth distorts politics. One of the libertarian things I like in theory is the idea of separation of economy and state, but in practice that's a brutally hard thing to achieve... much harder than separation of church and state. It is true from a social contract and legal point of view that money is not the same as physical coercion, and the state has a monopoly on the latter, but if you have loads of money you can buy the people who run the state and therefore obtain direct access to physical coercion. That's the problem. The super-rich can buy the government and then use force to maintain their status or pursue their own personal cultural, religious, or national peeves and hobby horses.

For the people I know, the biggest things are:

(1) Addiction engineering, and the fact that the tech industry found a way to transform what was supposed to be a machine to improve human education and cognition into a digital version of tobacco or a casino. I am personally disgusted by this. If you're working on a pop "social" app or a game and designing it to have a "compulsion loop," you are doing mind control and you suck.

(2) The fact that a huge number of people in tech have been "pilled" with hyper-elitist fascist-adjacent ideology. This one was a huge shocker to me as I watched it happen, especially because I read the texts and blogs that seemed to be driving it and was like "you people are smart... why are you falling for this?" I am by no means a hard leftist either, used to self-describe as a libertarian back when that meant what it sounds like it means. ... Then I realized that tech is full of nerds who (like me) got picked on in school, and that wound makes "you are aCkTuAllY a member of a cognitive master race" a powerful hook. It's not unique to nerds -- all humans are vulnerable to this pattern. You see it in groups of people who have been mass-abused through genocide or persecution for instance.

(3) A small number of people in tech have gotten rich to Gilded Age levels, and that burns in what I think is a stagflationary environment (we aren't calling it that, but for the "regular economy" we are basically in stagflation). If everyone was doing well and things like housing were affordable, I don't think people would care as much. I know personally I don't care if there are quintillinoaires out there somewhere as long as I am doing well and I can pursue my dreams... but if I'm not, it creates emotions of envy and hostility. This is natural hard-wired brain stem stuff.

Those are all rational reasons IMO and they all make sense.

I think the AI backlash builds mostly on #3, since AI threatens a lot of jobs in a stressed economic environment. The environmental stuff around AI (data center power etc.) is mostly exaggerated (except water in some areas), but its prevalence is driven by the fact that people want to find something to hate about this industry.

As evidence I submit the fact that artists, in my experience, are the most visceral AI haters. I get it. I do not believe AI is a threat to "true art." Not at all. But it is absolutely decimating the boring "potboiler" work artists used to use to make a living while they worked on the former, and it's always been brutally hard to make a living as an artist. AI also threatens programmers, but in that case we're talking maybe being knocked down from upper middle class to middle class not from lower middle class to bankruptcy.

(Total tangent -- I ran some numbers. If you telework 1-2 days a week, depending on your commute, you more than make up for using something like Claude Code heavily all day long. Same goes for skipping that DoorDash order. Cars use insane amounts of power if you really look, much more than data centers on an amortized per-user basis.)

Mobile is the UI/UX equivalent of… I don’t even know… a moon landing? A wonder of the world?

I’m not saying it can’t be duplicated. I’m saying if you want to build a mobile platform you need to approach it with appropriate respect for the incredible difficulty of making something that usable.

If they do that you’re using them wrong, or trying to use them as a substitute for thinking which they’re not.

You are the one in charge doing the primary thinking and guidance. The AI is an amplifier and a search engine basically.

Guns have only one main function. Phones have thousands.

Facebook was on balance good until the algorithmic timeline.

At the very least I think we should seriously consider taxing advertising.

It’d be almost impossible to pick and choose so just tax all of it. I think the knock on effects would be positive. It’s an almost unmitigated vice at this point and is the main engine behind this stuff.

The problem is not phones. Phones are fine. The problem is specific apps that make use of addiction engineering. These are bad on desktops too but the extreme portability of phones makes them a hundred times more potent.

Like all risks it doesn’t affect all kids equally either.

Some are less vulnerable for various cognitive reasons just like some are less prone to chemical addiction.

Kids with wealthier and/or more engaged parents or parents with more free time are also less vulnerable. Wealthier kids have more activities available and can often afford to have one parent stay home.

Lastly kids in healthier communities or suburbs or safe urban settings where they can roam free are less vulnerable.

They children of the poor, those with ADD or ASD conditions, and those with less third spaces or other activities are most vulnerable to becoming addicted to endless stupefying doom scrolling and addictive games that pre-train them for future gambling addiction.

It’s not just kids either. The elderly and the isolated become addicted to this stuff.

Addiction engineering is the problem, whether it’s via a phone, a web site, or a chemical.

IMO if you intentionally and knowingly engineer something for addiction you are committing a form of assault.

My experience is that AI is just that, a “mech suit for your brain.” It has no creativity or volition but has superhuman memory, superhuman speed, and superhuman context in some narrow cases.

So it takes a thought and unfolds it, looks up relevant thoughts and information, elaborates, works through implications, and in some cases can execute.

You could do all that but like doing math manually it would take forever. You could manually calculate a spreadsheet too.

I’ve toyed around with this hypothesis about dark music and edgy culture.

The roaring 90s were full of this: dark music exploring critical themes or things like depression and addiction, lots of social criticism, and edgy offensive humor. The mid to late 90s were an economic boom time for a broader segment of people and were optimistic in a way that’s hard to explain today.

I like to put it this way: it’s not that things were all better in the 90s. They weren’t. But people thought they were getting better.

Now people don’t want those things. They want light entertainment, either no criticism (the right) or only approved types via approved critical theories (the left), and offensive edgy stuff is out and may get you cancelled (which side cancels you depends on the content).

Dune isn’t a straight hero’s journey because Paul isn’t a hero, or at least is a tragic one. He’s not Luke Skywalker or Frodo.

I don’t watch many movies due to time but those I do watch tend to be odd, unusual, experimental, and sometimes polarizing.

A recent one I enjoyed was Rabbit Trap, a modern take on faerie lore. It was polarizing and got poor reviews.

Years ago I adored a two season series called Hellier. I have a hard time describing this. Ghost hunting show for thinking people? You have to suspend disbelief but it reminded me of Primer in terms of a lot achieved on little budget. If you want to see some hipsters try to use transcranial magnetic stimulation to talk to aliens this is for you.

I’ve enjoyed some horror for its tendency to subvert tropes: hero to villain, villain is really hero, everybody dies, etc.

But yeah. I can only watch blockbusters if I’m in a certain kind of mood where I want light entertainment with no thought.

The problem is the incentive structure. We could be doing good things with tech. The incentive is to use it to addict and destroy. Addiction and war is where all the money is, not education or health care or positive forms of entertainment and art. No money in those. Why?

I’m sure there was an early hominid version of this discourse. “Maybe bad to make sharp rock and sharp stick if this what we do with it…” “Mmm yes someday we make sharp rock big enough smash world.”