site banner
PaperclipPerfector  1mo ago (text post)   23888 thread views

Culture War Roundup for the week of August 3, 2026

This weekly roundup thread is intended for all culture war posts. 'Culture war' is vaguely defined, but it basically means controversial issues that fall along set tribal lines. Arguments over culture war issues generate a lot of heat and little light, and few deeply entrenched people ever change their minds. This thread is for voicing opinions and analyzing the state of the discussion while trying to optimize for light over heat.

Optimistically, we think that engaging with people you disagree with is worth your time, and so is being nice! Pessimistically, there are many dynamics that can lead discussions on Culture War topics to become unproductive. There's a human tendency to divide along tribal lines, praising your ingroup and vilifying your outgroup - and if you think you find it easy to criticize your ingroup, then it may be that your outgroup is not who you think it is. Extremists with opposing positions can feed off each other, highlighting each other's worst points to justify their own angry rhetoric, which becomes in turn a new example of bad behavior for the other side to highlight.

We would like to avoid these negative dynamics. Accordingly, we ask that you do not use this thread for waging the Culture War. Examples of waging the Culture War:

  • Shaming.

  • Attempting to 'build consensus' or enforce ideological conformity.

  • Making sweeping generalizations to vilify a group you dislike.

  • Recruiting for a cause.

  • Posting links that could be summarized as 'Boo outgroup!' Basically, if your content is 'Can you believe what Those People did this week?' then you should either refrain from posting, or do some very patient work to contextualize and/or steel-man the relevant viewpoint.

In general, you should argue to understand, not to win. This thread is not territory to be claimed by one group or another; indeed, the aim is to have many different viewpoints represented here. Thus, we also ask that you follow some guidelines:

  • Speak plainly. Avoid sarcasm and mockery. When disagreeing with someone, state your objections explicitly.

  • Be as precise and charitable as you can. Don't paraphrase unflatteringly.

  • Don't imply that someone said something they did not say, even if you think it follows from what they said.

  • Write like everyone is reading and you want them to be included in the discussion.

On an ad hoc basis, the mods will try to compile a list of the best posts/comments from the previous week, posted in Quality Contribution threads and archived at /r/TheThread. You may nominate a comment for this list by clicking on 'report' at the bottom of the post and typing 'Actually a quality contribution' as the report reason.

2
Top Bottom Old Controversial Comments
Jump in the discussion.

No email address required.

There is a great new Substack article about the populist backlash to AI data centers by Jasmine Sun. She also did an interview with Ezra Klein for those of you who like audio content.

I like the article because Sun speaks fluent Silicon Valleyish while also taking the time and effort to visit and talk to the people on the ground in the rural Midwest instead of condecending at them from the coasts. She asks an important question: why precisely do people hate data centers so much?

Looking only at the statistics, AI land use and water use issues are fake. Nowhere is running out of land, and nowhere (West of San Antonio at least) is running out of water. Are the people in these communities stupid? Do they not know how to shut up and multiply?

It turns out that a lot of the fear and anger stems from a combination of distrust and PTSD from deindustrialization. An interesting example is Janesville, Wisconsin, a town with a giant concrete slab in the middle of it where the GM plant used to be.

“Small communities don’t have the legal budget to negotiate well. You’ll never hold a company to a contract,” said Cathy, who feared Janesville once again tying itself to a single big employer—especially one in a “high-risk industry” like AI. City officials were promising to lower wheel and property taxes on residents, but that assumes everything goes right. What if the project gets abandoned? How does Janesville avoid “another GM plant scenario”?

Fascinating. The collapse of the auto industry in 2008 was apparently so bad that it is better to prevent industrial development alltogether in order to avoid the risk of another collapse. It almost makes sense until you think about it for 30 seconds.

"I ask the three what they envision being built here instead, but don’t get much of an answer. I suspect that Adam is partly right—maybe this is a choice between a hyperscale data center and a concrete slab. And Alyssa, Cathy, and Kaycee would rather pick the slab."

Some of the objections Sun hears in other towns fare a bit better.

In 2017, the Village of Mount Pleasant, Racine County, and State of Wisconsin offered over $4 billion in tax incentives and infrastructure investments to win a bidding war for a Foxconn plant. In return, Foxconn promised Mount Pleasant—population 26,000—a $10 billion investment that would create 13,000 jobs making flat-screen TVs. The village eminent domained the land and prepared infrastructure. But Foxconn kept downsizing their plans, citing high labor costs. By the end of 2024, just over a thousand jobs had been created. Mount Pleasant was saddled with over $183 million in debt that Foxconn and Microsoft are still servicing.

The memory of the Foxconn fiasco is still raw. We heard it referenced in conversation after conversation as a flaming symbol of corporate failure.

This is the steelman of the anti-data center position. People are worried that the big companies proposing data centers are going to Foxconn them and not deliver the supposed benefits.

My conversations make me wonder if the technical content of data center deals is mostly beside the point. Closed-loop cooling systems? “I don’t believe them.” Paying millions in taxes? “I don’t believe them.” Creating a thousand jobs? “I don’t believe them.” I hear a reflexive skepticism of every claim.

Rather, the way that AI companies engage with communities—forcing NDAs, dangling billion-dollar promises, pushing environmental externalities far away from AI’s wealthy user base—resembles a classic story about dark money in politics. Your politicians are being bought by billionaires and gigantic corporations to screw you over. It’s not Skynet: it’s the oligarchy.

One of the things that comes across especially clear in the Ezra Klein interview is just how unprepared the AI companies were for the backlash. They didn't realize how sketchy it comes across to insist on NDAs with local officials when negotiating deals. They thought, as a straightforward application of Coase's Theorem would tell you, that any bumps in the road could be ironed out by handing over a big bag of money. It turns out that people don't work that way. It feels like a bribe, and Americans do not like being bribed.

She asks an important question: why precisely do people hate data centers so much?

The article seems to be either eliding something important, or treating small-town-folk as serious rubes -- while I'll grant that normies are not likely to be as familiar with the details of AI-doom discourse as the average Mottizen, surely it trickles into their consciousness at least a little bit?

Which is particularly relevant because it's being promulgated by the same people trying to build the data centres -- it's similar to the dust-up with fedgov on "AI safety".

Only instead of:

"Hey we are building this technology that can hack everything and take over the world, you should give us money -- but we need you to leave us in complete control and not regulate us in any way"

it's

"Hey, we are building this technology that will take over not only your job, but the entire world -- also there's a strong chance that it will lead to the extinction of humanity. If you help us out on this, we will happily... pay municipal taxes at a somewhat reduced rate(!?) as negotiated by our slimy lawyers"

Does this seem like a good deal to you?

Speaking out of both sides of one's mouth and then being shocked (shocked I say!) when people catch wind of what you are saying to others and become uncooperative is a strong sign that one is not as smart as one thinks one is -- probably a pretty big bear signal for these companies if you ask me. (doesn't help that both sides of the mouth sound suspiciously like an evil robot)

Small-town bumpkin real-estate developers have been able to mint money by working municipal politicians into approving terrible deals (in terms of, say, servicing costs for new subdivisions) for many years -- you do it with some combination of bribery and 1-1 marketing; not sure which of these Sam et al are failing at, but if they can't figure it out they definitely deserve to lose.

One thing about data centers I keep hearing is that they don't meaningfully contribute jobs to the local community, and I'm not sure why this would be, relative to any other commercial project of similar scale in a given community. Even if the compute on these servers are being used at some arbitrary location on Earth that's completely unrelated to the data center, that specific server needs to be installed, maintained, upgraded, replaced by someone who is physically there. In the long run, sufficiently cheap AI- or remote-controlled robots could replace this, but we're not there yet and I don't think that many people think it's around the corner either, like how so many people seem to believe AGI or even ASI is right around the corner. That person who physically checks on the servers and fixes or replaces it when something breaks or upgrades it just when faster hardware is sufficiently cheap and available to make economical sense would almost definitely live near their job at the data center, and so either that person would be a local or a local import who would spend the money the data center company pays them buying stuff and services from local shops. This seems exactly analogous to any other job of e.g. being a baseball stadium clerk or a warehouse worker.

Yet data centers get talked about as if they're just big slabs of compute that these AI companies build and just place somewhere for other people to be bothered by, never to require paying humans to actually maintain or upgrade the thing. Do these data centers require less jobs or less paying jobs on a per-dollar-invested basis than other similar commercial projects to such a great extent that they can be rounded down to zero? If so, what if we mandated that data centers provide make-work jobs to the local community where people are paid to stand watch to maintain computers that need 1000x less maintenance than they can provide? I've sometimes thought that, given the sheer complexity of the code and tools AI is capable of generating, which far outstrip the ability of any human or even possibly team of humans to understand, and the fact that AI lacks agency or the ability to be held responsible, every company in the future would eventually have precisely one and only one employee, whose job is just to sit around watching the AIs that are literally incomprehensible to them and be summarily fired and replaced if anything goes wrong as a result of AI behavior. Give them an official job title of "scapegoat." We could start doing something like that with AI data centers (give them all golden parachutes so that the money flows to the local economy regardless).

whose job is just to sit around watching the AIs that are literally incomprehensible to them and be summarily fired and replaced if anything goes wrong as a result of AI behavior.

Relevant SMBC

Contrast AI data centers with more traditional industry, say a car factory (perhaps from 20 or 30 years ago).

Per million dollars invested, the car factory created tremendously more local economic activity. I imagine they might have multiple full-time jobs per million dollars, welders and mechanics and office workers who would in turn go to cinemas and super markets and so forth.

In AI, 1M$ buys you perhaps 25 H200 GPUs. They probably fit within a single 2m rack. You install that rack, along with thousand similar ones, and send a technician to replace failing fans. I don't think that this is the price segment where you care about saving a few dollars on fans, so it seems entirely plausible that your 1M$ investment will never require a technician during its economic lifetime. Sure, there is some overhead, like power supply and cooling, but this likewise is stuff which can be made reliably if you are willing to spend big bucks on it.

The intellectual labor of running the data center (to the degree that it still involves humans) is not local at all. The nerds sit in SV and do not spend a dime in the local economy.

For a community, the liability of an industry is proportional to the investment. Permitting someone to build a small theme park for a few million dollars is a limited risk, at worst the town can cough up the costs and shut it down. With a billion dollar investment, things look different. Once that data center is built, it will run, come hell or high water (though paperclips or the bubble bursting might still cause it to be shut down), because the investors have billions on the line and will be sure to phrase the contract in such a way that whomever is running on a "cancel the datacenter" platform will not be able to do so.

If the datacenter offered a thousand local jobs, that might still be worth it. But it does not, it offers a handful of jobs -- plus some temporary economic surplus while it is being built.

Furthermore, the locals likely have an acute awareness of how likely their elected representatives are to actually make a great deal instead of a complete blunder.

From their perspective, it may be fairly described as a Faustian pact -- you do not know that the demon is going to fuck you over, but you know very well that he is much smarter than you are and does not have any straightforward incentive not to fuck you over, so you do the math and decide that you will pass.

What exactly is the risk that the local community is taking on here? There are no toxic chemicals in a DC. It doesn't make any noise and is very unlikely to catch on fire (it would ruin all those fancy GPUs). It uses some power from a national grid that the local community has very little part in. It uses some water but probably less than the farm it replaced. Essentially, it's an inert concrete and rebar box that minds its own business and pays taxes (incl. local ones).

I'm not a big AI fan. I think it's a giant bubble that's going to collapse some day leading to some ugly financial consequences. But I still don't get all the fuss around data centers.

Loudoun County, Virgina is the data center capital of the US. There's 50 million square feet of data centers built or being built there.

Despite every hyperscaler, data center leasing company, and major tech company having at least one and often multiple data centers there, only Amazon appears on a recent top employers list in the county. And Amazon operates more data centers there than anywhere else in the US and has significant management of other data centers from there.

Loudoun County only claims 15,000 jobs in the data center ecosystem (meaning not just operations but the jobs associated with building them, too).

An older study (from a time when only 30 million square feet in were in operation) lists 5400 jobs actually operating those data centers and 65% of those jobs being outside the county (no break down of fully remote or commuting from a neighboring county).

Compare that with Witchita Kansas which is a major location of aviation assembly. Sprint Areosystems has a 20 million sq ft assembly campus in Witchita, which employes 16,000 directly.

That's 3x the direct jobs, for 2/3 or 2/5th of the space. If most of those non-county jobs are fully remote, that's more like 10x the local job density between advanced manufacturing and data centers.

You're right in general, but some of it is just that high paid tech workers don't want to live in Loudon County over the other nearby areas. My prior job was for sw infra engineering work on a (first party, non-AWS/GCP) data center in that area, where physical proximity to the building didn't matter for my role but did for people on teams I worked with every day. Our office was in Herndon (Fairfax County, right on the other side of the border), and I lived with the other yuppies in Arlington.

Yeah, but realistically most places don't have the choice between an aviation assembly plant and a data center. They've got the choice between a data center and some sort of unused land.

Sure, the contrast was just to illustrate the relatively low job density.

The local issue is more about how much tax incentives get offered and how much electrical infrastructure and how well area generation can supply the new electrical demand. Sometimes land may be better for everyone except the local politicians.

The "choice" in a local planning sense for a municipality choosing where to zone for AI, if you have a site for a 1,000,000sqft industrial building, is between a data center that might have two dozen or so permanent employees and a warehouse that would support a thousand unskilled jobs or a factory that would have 800 skilled workers.

So yeah, rounds down to zero.

In the end datacenters don't make long term jobs: to the town they're in, they're essentially inert boxes that pay taxes, similar to solar farms and many other low labor projects.

The thing is that the citizens instinctively know that if the government gets a huge windfall, $0 of that will go to helping the citizens, and it will all get wasted. I know a certain town baleeted its public transit to save 1c of sales tax, and absolutely none of the savings went to the people. The mayor spends it all on a bunch of vanity projects.

that pay taxes

Even that's not a given.

The Shapiro administration’s latest budget estimates show Pennsylvania could lose out on about $2 billion in revenue by mid-2031 due to a tax break that some lawmakers want to repeal.

Sometimes I wonder if all of the problems with modern politics are tied to cheap transportation and communications technologies. People care about status. They want people to like them. Noblesse Oblige is founded on making the people like you. If public officials lived in their communities and interacted with people in their communities, then doing good deeds for the people and saving them a bunch of money would be a vanity project! The status and validation and pride that the mayor seeks for themself ought to be from the people. But if they just fly around on jets talking to rich people and politicians from other cities then the status turns into big ostentatious displays that rich people care about.

Somehow we need to ground politicians, their status, and their emotional state to be related to the people. If the majority of the time they spent socializing was with random people from their districts then they would automatically seek to make the people happy because that would be what flatters their egos.

The most recent actual news on data centers is that Greg Abbott’s data center plan is polling at +70.

For those who aren’t familiar, his plan is basically ‘let them build data centers, but publish data on it’. People don’t trust Silicon Valley nerds to be honest, that’s why there’s all this backlash. It’s entirely possible to invert an 80-20 issue by just framing it at ‘someone accountable to the people is checking the deal’.

Texas is an interesting case there because ERCOT has long-term (2032) forecasts expecting peak grid demand to rise from 91GW (last month) to 370GW. More than 4x! I'm frankly not sure where they expect energy to come from, but solar (and battery storage), wind, and natural gas have been growing pretty fast for years now. But probably more important is that someone is at least thinking about the distribution infrastructure to do so.

how unprepared the AI companies were for the backlash

They don't see themselves as the grifters / baddies. High on their own supply.

The song almost writes itself.

[Lyle Lanley]
All right, folks, step right up and I'll show you my idea. I give you... the Springfield Hyperscale Data Center!
I've built 'em in Brockway, Ogdenville, and North Haverbrook, and by gum, it put those towns on the cloud!

[Lyle Lanley]
Well, sir, there's nothing on God's green earth like a genuine, bona fide, liquid-cooled, hundred-megawatt data center!
What'd I say?

[Ned Flanders]
Data center!

[Lyle Lanley]
What's it called?

[Patty and Selma]
Data center!

[Lyle Lanley]
That's right!
Data center!

[All]
Data center... data center... data center...

[Miss Hoover]
They say those fans hum day and night...

[Lyle Lanley]
That's just the cloud — it's soft and light

[Apu]
Won't it drain our water table?

[Lyle Lanley]
Closed-loop cooling — nice and stable

[Barney]
Any jobs for brain-dead slobs?

[Lyle Lanley]
Six whole full-time server jobs!

[Lisa]
What'll happen to our power bill?

[Lyle Lanley]
The AI pays! ...Well, someday it will

[Abe]
This sounds like the work of the devil!

[Lyle Lanley]
No, sir — my uptime's on the level

[Wiggum]
My cop show keeps on buffering, man...

[Lyle Lanley]
Free town gigabit's in the plan!
The future calls — you've got no choice
Log on, look up, and raise your voice!

[All]
Data center!
[Lyle Lanley]
What's it called?

[All]
Data center!

[Lyle Lanley]
Once again...

[All]
Data center!

[Marge]
But the power grid's half-broken...

[Bart]
Sorry, Mom — the shareholders have spoken!

[All]
Data center!
Data center!
Data center!
Data center!

[Homer]
Data... d'oh!

The collapse of the auto industry in 2008 was apparently so bad that it is better to prevent industrial development alltogether in order to avoid the risk of another collapse. It almost makes sense until you think about it for 30 seconds.

Think about it for longer than 30 seconds. Remember AOC among others successfully stopping the new Amazon headquarters in New York? Why would they do that? Didn't they want economic development and jobs?

How places attract such large mono-developments (and God knows, Ireland has led the way in this) is by promising very, very favourable conditions. Tax cuts. Subsidies. Grants. Where does that money come from? The local community. New York might be big enough to absorb those costs, but the locals didn't think so when it came to Amazon and the subsidies held out as bargaining chips. Janesville certainly can't, and what about if the local government commits to the new data centre, things go forward - and then halfway through for some reason it gets abandoned? No jobs, locked in to contracts that require penalties (probably) and more crushed hopes (and more than hopes, if people have moved back home in anticipation of getting work on construction etc. for the new site).

I'm living in a town where we now have, in effect, one remaining major employer. Over the last three decades or so, branch offices of national companies have closed down or moved elsewhere, and the remaining major employers amalgamated/got taken over by other companies and the work was shifted elsewhere. I can sympathise with the people of Janesville being afraid that this is one more mirage that they will end up paying for when all the glittering promises pop like party balloons the morning after. It's not "would prefer a concrete slab to a hyperscale data centre", it's "would prefer not to put all their eggs in one basket".

Amazon and data centers are net payers of taxes. If AOC and New Yorkers selfishly wanted more tax revenue, then they'd approve of the Amazon headquarters. I reject the premise that data centers or Amazon headquarters are tax drains.

Loudoun County, VA approved of those data centers and they pay around half of all property taxes in the county to the benefit of the people who live there. Of course they have to pay their taxes.

Of course, AOC apparently didn’t recognize that the tax incentives weren’t something they’d get to keep if they didn’t approve the deal- they were an offer to just not collect some of the taxes that would be owed, but aren’t due to Amazon not building.

Tax cuts. Subsidies. Grants. Where does that money come from? The local community.

Almost all of these deals bring more tax revenue than the government services they consume, after accounting for all incentives.

Yes, but for the average taxpayer, all they see is yet one more line item on their taxes and the supposed benefits often never really materialize. What do I care that a line on the city budget goes up if that doesn't translate to improved roads or schools? That increased tax revenue never gets passed down in the form of a rebate or lowered residential property tax. It just gets funneled into more stupid shit I didn't want the city to be doing in the first place.

I'm not anti-data-center as a general idea. But somehow every data center plan in my town is absolutely moronic.

My particular town has one main commercial corridor. On this corridor you have stores, movie theaters, banks, offices. People moving back and forth across it throughout the day. Now there used to be a major company headquartered there, they since moved to new offices down the road. Developers want to convert the old offices into data centers, a few million square feet.

My first thought upon looking at the plan was, you'd never do this in Sim City. You have one of your most prime pieces of commercial real estate in the region, and you want to turn it into a box that no one uses and nobody goes to. You want to give away that access, that site with easy walkability and simple drivability to literally everything in town, and turn it into a box that hums. You're going to take this prime real estate away from people.

It's just a stupid development choice. And I get that the market doesn't owe me Sim City type optimal decision making, but it's a weird thing. I'm not anti-data-center, but there's obvious good places and obvious bad places to put them, and so many of the choices made seem to be obvious bad places.

Data centers need to be beautified. Have competent representational artists and classical architects work on the facade, so they don’t look like an ugly blight on downtown. Make them a thing people don’t mind seeing.

Better still: large forested buffer around the building, reduces noise and makes it invisible from street level.

My father strongly believes that the future of data centers is in placing hydroponic greenhouses around them.

Time to bring back the Frutiger Aero style of thinking, I see.

Hmm. A palatial botanical conservatory with orchids in the dead of northern winter at least sounds like a civic use of waste heat. Maybe Milwaukee should relocate the Mitchell Park Domes to Mount Pleasant?

Splendid colonnades and grand entrances with beauteous manicured forests stocked with parrots and wallabies… yes, thé people will appreciate.

It does seem to me that the most parsimonious explanation is that the developers involved in this are mostly treating it as a gold rush rather than coming in with well-thought-out plans. OK, maybe it's cheaper to convert an old company headquarters into a data center, but it's also certainly understandable that people don't like it. Not helped by the AI companies' intentionally terrible PR strategies.

And you'll get similar results, as well.

One of the claims I've seen for the benefit of data centers is, well, jobs. As someone who has friends looking to get into Data Center construction, given from what I've seen, the timeline for construction seems to be around 2 to 3 years for each new data center.

In the local area, for those 2 to 3 years, the numbers will probably be rather rosey - you'll get a glut of new people spending a fair amount of cash from things from groceries to restraunts to RV parks to hotels - business is booming! Numbers go up!

Until they leave.

Whoops. So now businesses that were operating under the assumption of a cash flow that no longer exists - and as someone who's watched more businesses fuckup due to accounting budgeting and making really stupid decisions than I care to admit to, oh boy is this the likely outcome - if they can't react to the economic shock, more than likely they'll go under.

So now you've got a data center likely set in the most moronic of location and a bunch of businesses now no longer able to make ends meat because the circumstance under which they were flourishing no longer exist.

And you've still got the fucking data center.

Do I have numbers to back this up? Studies? No. All I'm going off of is a little bit more economic knowledge than the average person and watching businesses do really stupid things during covid. (Businesses got flush with cash during Covid due to government grants to stay open. So they planned around that sudden cash flow. Guess what happened when it dried up?)

Putting all that aside, even people who are pro-data center have no issue pointing out that the entire PR look is bad. And, in my personal pet peeve, I'd seen more than a few people try to compare anti-data center people to anti-nuclear power, nevermind how moronic such an arguement is.

One of these days, I'd love to see a social study to find out what causes people's brains to flip and do the meme.

And, in my personal pet peeve, I'd seen more than a few people try to compare anti-data center people to anti-nuclear power, nevermind how moronic such an arguement is.

What's moronic about it? It sounds very easy to me. Nuclear could have been the next generation of energy but was constrained by regulations protecting the places where they would be built. "Data centers" could be the next generation of tech but are threatened by a movement that wants regulations to protect the places where they would be built. And in both cases it scans really simply as all mixed with more than a little bit of Luddism.

Nuclear you can point to solid returns that would benefit the common consumer directly - less reliance on hydrocarbons, energy independence, cheaper energy.

Data Centers you point to with the returns that basically boil down to 'Big Businesses need it, LLMs are magic bro, trust me, we've got to beat China!'

So color me a skeptic. On Data Centers, not Nuclear. For fuck sake people, just use the magic rock and be done with it.

Data Centers you point to with the returns that basically boil down to 'Big Businesses need it, LLMs are magic bro, trust me, we've got to beat China!'

> is it a strawman or an argument
> "it's a good argument sir"
> look inside
> it's a strawman

We wouldn't be having this conversation right now without data centers.

In my day we called them R/C helicopters server farms

Extraordinary claims require extraordinary evidence.

You(and others) are making comparisons that Data Centers are just as important(if not more, from what I've seen people say) than Nuclear Power. And when I go digging around, that's the answers I typically find.

Brother, you accuse me of building a strawmen when all I'm doing is rooting around in your hayfield, looking for a point. Hell, you haven't even offered up a rebuttal. If you've got something in your pocket, I'm all ears.

I asked you a question about something you said, and you proceeded to defend a different statement. You then claimed that data centers do not have “solid returns that would benefit the common consumer directly,” and that nobody can come up with any. Now that I point out that this is an obvious strawman, you become hostile accuse me of bad faith.

This is ridiculous.

This is besides the fact that it appears you appear to misunderstand simile. At least, when people say “opposition to data centers is like opposition to nuclear,” it is not relevant for you to assert that nuclear is more important than data centers.

I’m happy to continue this conversation, but I’m confused as to whether: 1) You want to defend the notion that AI is useless 2) You want to defend the separate notion that AI is less important than nuclear 3) You want me to introduce some other opinion so you can argue against that instead of having to defend your own contradictions

More comments

I think the AI bubble is uniquely confusing, because the real estate pros building them out are treating it like musical chairs and the companies commissioning them are treating it like the Rapture. Real estate pros figure it's a bubble, in the sense that eventually they will be overbuilt and stop building, so if I want to cash in on it I need to do it right now before the music stops I can't wait around. AI companies treat it like the Rapture, it doesn't matter if they overspend now because AGI is right around the corner and then money doesn't matter. They're unconcerned with the idea that the music will stop, they're concerned with controlling the playlist because whatever song is playing today will be on repeat forever. Combine the two and you're getting a toxic bubble.

It's also a problem in my town because of prior government interventions to subsidize the construction of new office buildings, creating a glut of new office space, which in turn makes building office space impractical. So you have sites that would be perfect for an office park being turned into data centers.

the developers involved in this are mostly treating it as a gold rush

Exactly this. Happened most recently with us during the Celtic Tiger when developers were embroiling themselves in millions upon millions of debt for landbanks and when the construction bubble burst, all that land was worth nothing and they went under (and the banks that lent them that money had to be bailed out, and we had to go begging to the IMF and the EU for loans to help us bail them out).

insist on NDAs with local officials

This is insane.

That is typical. Private details of as-yet-unfinalized negotiations are to be private. Once finished they will publicly announce the outcome.

Whether typical or not, it should not be allowed. There are occasions when a public official might have to do things in secret for the good of the public (military matters and so on), but deals like this are not among them. And whenever government affairs can be conducted publicly, they should be.

The only reason for NDAs in this context is to prevent property owners from coordinating to obtain fair market value for the land the companies are trying to buy. It's for the sole and express purpose of transferring wealth from locals to AI companies. There is no public interest excuse.

Is this not fairly standard for a large enough (and sensitive) deal?

Why not both?

I can't speak to how standard NDAs with public officials are, but they are definitely insane. The governed can't consent to what they aren't aware of!

Fascinating. The collapse of the auto industry in 2008 was apparently so bad that it is better to prevent industrial development alltogether in order to avoid the risk of another collapse. It almost makes sense until you think about it for 30 seconds.

That attitude makes sense to me, assuming you think a crash is likely in the short term.

If a place is reliably poor, people can plan around it. Young people can move to other cities where there is more work and those who stay can plan for a smaller lifestyle.

If the town builds up high around money coming off a new data center then there will be significant damage if a crash happens. There will be people stuck with mortgages they can't pay off, lots of businesses going under, and families who have put down roots facing the choice between moving and having no job.

And the jobs never materialise. Big company comes in, claims it will create X hundred new jobs over the next five years, five years later it's around 100 jobs if at all.

I think data centers have the potential to benefit the municipalities that they build in. Some jobs, upgraded infrastructure, lower prices. But do I trust every municipality in the US, great and small, to competently and non-corruptibly negotiate with sharks from trillion dollar companies? Haha no, I'm sure some municipalities will get fleeced or sold out.

That's probably the biggest source of anxiety even though on net they should be a benefit.

Do you think this concern is silly? I'm not sure what your opinion actually is?

Surely there's a town with a data center they can go ask how good it was for them? Or do they all have NDAs too?

Seems to me to be a perfectly solvable problem, just show the value. If they can deliver jobs and tax money to localities, they will be welcomed. If it's all a pyramid scheme to saddle local governments with tech companies' expansion costs, that will come out too. Just takes some time.

They do in fact deliver tax money. Around half of property taxes in Loudoun County, VA are paid by data centers. That county allows building of data centers and reaps the rewards.

I guess the idea is for journalists to actually go there and ask the residents whether their water is now toxic sludge and whether their electricity bills are now in the vicinity of gorillions of dollars as the activists say.

Surely there's a town with a data center they can go ask how good it was for them?

Are these new datacenters really that different from the ones we've been building for literally decades? AWS and Google have had giant server farms longer than anyone was really looking at machine learning like today. Heck, I remember people complaining about the NSA's data center in Utah slurping up all our emails, not its water usage. Complaining about this seems a new idea, but I don't recall too many complaints about the facilities previously. On the other hand, maybe it's like fracking, which wasn't new a decade ago, but really took off at scale.

On the gripping hand, I'm really not sure who's paying for all these new data centers, how they intend to pay the bills if AI tokens are fully commoditized, or why they're being put close to neighborhoods. AWS has existed in multiple locations (they don't disclose exact addresses, but it's probably not a single building) for a long time, and doesn't seem to have garnered quite the same level of complaints.

The water usage thing started because Karen Hao's book Empire of AI, made a unit error (or "error") which claimed data centers use 1000x the water they actually use. Then various propaganda groups (many likely Chinese Communist Party funded) started jumping on it. Often they claim the water use of the "data center" as including an imputation of water use by the power plant which produces their power, while also claiming the power use as a separate downside.

It's certainly true that a municipality could get fleeced in some way if they provide tax breaks or other incentives for a data center, but that's true with any large project. Data centers aren't special.

All data centers have a mix of computation, data storage, and network connections in different proportions.

Netflix requires massive amounts of bandwidth to serve streaming HD video, so their data centers have (relatively) oversized fiber optic cables connecting them to the rest of the internet. YouTube has tons and tons of data relative to how much it's viewed (how many YouTube videos have <100 views? How many Netflix ones do?), so they need relatively large amounts of data storage.

AI is about as close as you can get to pure computation1 (not much data is stored, and sending text at human-readable speeds is utterly negligible), so their data centers have modest bandwidth and storage, but massive amounts of processing. That takes more power and therefore cooling than other types. Claude found around 1-2 MW of power consumption per acre (of the full site, not just the building footprint) compared to about 0.1 MW/acre for heavy industry like smelters or 0.3-1 MW/acre for other big data centers.

1 Crypto miners are apparently even more computation-heavy relative to bandwidth and storage, but they aren't exactly popular either.

IIRC YouTube spends a surprising amount of processing power transcoding uploaded video, to the point where Google has developed custom ASIC hardware to do so.

I guess I don't know how AWS lays out their physical locations, but "all EC2 hardware in one building" would extremely computation heavy, whereas S3 would not be. But perhaps they're intermingled for some reason. AWS and Google also have a lot more institutional experience building these things, which probably smooths out some of the wrinkles.

For superscripts on TheMotte, the "^" markdown doesn't work, but you can use HTML:

<sup>$SOMETEXT</sup> => $SOMETEXT

Are these new datacenters really that different from the ones we've been building for literally decades?

A bit different. They use significantly more power and they're being built in such a hurry that the classical hyperscaler commitments to doing it carbon neutrally or with highly sophisticated cooling that leverages natural advantages and keeps power draw down are being dispensed with.

A famous example is the one Elon built in a hurry where they placed a ton of natural gas generators on site because the local grid couldn't supply them, possibly in violation of environmental impact rules.

Pollution from Musk’s unpermitted xAI power project hits hardest in Black communities

Smells like political groups fishing for ways to bother elon

Regardless of how many black neighbors, towing a bunch of portable gas generators on site to get your GPU farm up and running in months is still a huge difference from how hyperscalers used to try to build out near hydro dams and negotiate decades long power purchase agreements, with credible attempts at carbon neutrality.

credible attempts at carbon neutrality.

You mean buying carbon credits from the lowest bidder?

No, I mean the opposite of that. Like locating near grid connected hydro-electric power instead of wheeling gas generators on site, and doing capital development to exploit natural cooling instead of just installing more CRAC units. Shit that takes a lot of engineering effort and time. The classical hyperscalers took this extremely seriously in the 2010s. This used to be a point of real pride at Google, and having come from working on finance datacenters before that it seemed like beaming up to the Enterprise.

In the AI lab era they're in too much of a rush to care anymore. Which is what the grandparent comment was asking about.

What else can they really do? There's roughly zero surplus hydropower sitting around waiting for a customer, and everything else has a lead time of double digit years or is even worse in terms of nuisance (E.g., massive diesel gensets).

I think Meta must have anticipated there being some opposition over a decade ago because we’ve near always coupled datacenter construction with big charitable projects. The earliest I can remember is Forest City, NC. Not only did we present the datacenter itself as bringing well paid jobs to one of the poorest communities in the state, but I remember us doing grants to nonprofits, donations to schools, and providing free broadband services.

The company puts out press releases but I’ve never seen anything about these programs in a news article.

Meta is actually one of the best at trying to be responsible and taking significant measures to address peoples' concerns, but for whatever reason they don't really talk about it much and certainly none of it gets reported in the media.

I haven't read the Substack article yet, but I did read the Klein interview, and what stood out to me was this:

And I went in and asked these organizers: Do you guys use A.I.? Do you find it useful? This was one of the top questions that my friends in San Francisco wanted me to ask: Are these people using ChatGPT, and they don’t even realize that the data centers are how they can use it?

And what I found was that a lot of these folks did say: Yeah, I’ve used it to draft an email or make a meme. They’re not denying that A.I. might have any possible utility at all, but they clearly didn’t see it as essential in the way that cars, energy and housing are essential.

They clearly saw it as like a widget — a toy — and maybe there are these risks, maybe there’s the job stuff, but fundamentally they were like: This thing is not that useful. I don’t really see in my personal life how this could justify these gigantic valuations.

And so I was talking to, for example, Charles Franklin, who runs the Marquette Law School Polls in Wisconsin, and he was explaining that usually you see 50-50 polling on issues where you have a strong anti case and a strong pro case. And the only reason you’re seeing this 70-30 bipartisan opposition to data centers, no matter whether you live near a data center or you don’t — which means it’s not just NIMBYism, it’s you don’t want a data center in anyone’s backyard — is because there is no strong pro argument for it. There is not even a fight that’s really going on.

“Nobody wants it” was a phrase I heard over and over.

A big part of the dissonance, it seems to me, is around the perception of what 'AI' is and whether it's valuable. Clearly a lot of people in Silicon Valley believe that this is a massively transformative technology that will be vital to the future of the country if not the world. They're frequently either outright singularitans, or simply believe that it willl turbo-charge the economy, massively increase wealth, change everything about employment, and so on.

But these people, and I admit I think they sound a lot more sensible than the tech people online, actually use the technology and go, "meh". They see it as a mildly amusing toy. Maybe it has some uses, in some niche fields, but for the most part it's just not very interesting. If they didn't get access to it, they would not particularly care.

An auto factory - you at least understand why cars are useful, and why the factories need to be built. A data centre is a huge investment for a thing that... isn't useful. All of this is in aid of a technology that's just fundamentally pretty crap. No one wants to make sacrifices for the infrastructure to produce more LLMs because no one thinks that LLMs are worth sacrificing for.

There is the marketing issue as well, of course. Pro-'AI' people tend to sound very cult-y online, at least in my experience. Many of them sound like lunatics. A lot of them also, well, to be blunt, are un-charismatic nerds. Sam Altman, Dario Amodei, Elon Musk, Mark Zuckerberg, etc., are not convincing ambassadors for a technology. I can understand nobody believing their promises - none of them project credibility, or even really respectability. The 'tech billionaire' is a cultural archetype at this point, and it is an archetype that one associates with immaturity, frivolity, narcissism, and callous disregard for other people and for human value in general. Last week I was talking about the book destruction story and it fits with the image again, as if there's this small cult-like group of anti-human tech-heads eager to sell the whole inheritance of the human community for digital pottage. I understand, in my gut, the feeling of disgust.

They clearly saw it as like a widget — a toy — and maybe there are these risks, maybe there’s the job stuff, but fundamentally they were like: This thing is not that useful. I don’t really see in my personal life how this could justify these gigantic valuations.

Yeah, that's the thing. All the proselytising for AI that I see - and that includes people discussing it on here and giving examples of what they use it for - is nearly all "it's great for coding". Yes, and I'm not a software engineer, have never coded in my life, and used Copilot once where, admittedly, it was useful. But in my day-to-day job, I don't use any AI and don't really as of yet see any need for it.

And most people are like that. "I work as a school caretaker. I'm a childcare practitioner. I work in a supermarket. I'm a nurse's aide. I'm a panel beater. What is this amazing difference AI is going to make to my life, make me 100x more productive, and give me $$$$$$ in stock options as my employer's valuation goes to the moon?"

Now, if we get proper business-oriented AI, not the "just use it, everyone says it'll make us a zillion times more profitable, no I don't know what you can use it for" versions being pushed now, then that will change. But right now it's "don't need to hire junior coders, our senior guys can do five times the work now" and "maybe office work? maybe?" for work, and mostly "ChatGPT for your online buddy/therapist/girl or boyfriend". That is the major use for ordinary people, I think: AI as 'my online best friend and lover'. And if you're average enough that you have a real world girlfriend or spouse, you don't see AI as being any good to you.

That's certainly the pattern I've noticed in the replies to me here. People who work with data think that LLMs are great and the wave of the future. They automate so much, they increase productivity, and so on. Laptop people, so to speak, think the tech is valuable. People who work with other people or with the world, though, still don't see the use. LLMs do nothing to help in roles that are primarily about interacting with other human beings, or which are primarily about interacting with physical stuff.

I wonder if there's a level of myopia here, where the data people assume that their kind of work is the whole world. Then again, to be fair, you could make the case that I, with my people-centred job, think that people work is the whole world, so we are both blind to a whole section of society and economy. On the inside view, I'd defend myself by saying that statistically, people and physical stuff do in fact make up almost all the economy, but on the outside view, should I trust myself?

As a laptop person, I’m confident LLMs are useful, but under no illusion they’re useful for everyone and everything. Not even for myself, because I don’t in fact just sit and work on my laptop all day.

Likewise, even very non-technical people can benefit from LLMs, for example by placing an LLM between themselves and the modern bureaucracy (forms, customer service complaints, legal issues, medical diagnoses etc.).

by placing an LLM between themselves and the modern bureaucracy (forms, customer service complaints, legal issues, medical diagnoses etc.).

Do you want to drive my blood pressure up? Because the chatbots replacing customer service agents are totally useless for dealing with customer service complaints, and most websites make it extremely difficult to find a way of contacting them to complain or cancel contracts.

Maybe an LLM can do all that work but in the end it'll still be me on the phone trying to find a human to solve the problem for me.

I'm talking about you sending an LLM to talk to the LLM customer service, or work through the convoluted website.

Even worse layers of obfuscation.

Right now my internet is down and after logging the complaint, it will take seven days for them to send out an engineer to look at the box on the telephone pole outside my house. No LLM is going to make that experience easier for me.

I'm not a programmer either, and I do get annoyed at the way 90% of AI discussion is centered around coding, but come on. Software is eating the world. Anything we do, we do with software. Cars have software, TVs have software, phones have software, books are increasingly read in e-ink readers, etc. Software is as fundamental to our life as oil, or electricity. Imagine a technology that made oil workers 5x more productive, and thereby made oil cost 1/5th as much. It would not just be about how much you save at the pump; it would mean cheaper groceries and plastics and plane tickets and probably entire new applications that nobody has ever implemented because they did not make sense at the previous price level. That's what AI is.

My vintage Land Cruiser doesn't have software, nor does my record player, my guitar, my film camera, or my porch swing.

And you know what? I'm a helluva lot happier when I'm using any of those things that I am using just about anything with software.

Given the way cars are going, I'm inclined to say that having software is a downgrade for my car.

Markets: I consent

Investors: I consent

Isn't there somebody you forgot to ask?

In the UK we don't really have the same data centre showdowns described in this article, because the UK is already such a planning hellhole that no one would bother. Battles tend to be a lot smaller, focused on random businesses and housing developments. Nonetheless, so many of the NIMBY arguments seem to apply regardless of scale, one of the biggest of which is some kind of weird, communist, demand-management approach. "We don't need that here". "No one wants that". "Why are they building X instead of Y?"

Well, who the fuck cares about your market opinion, random local? Why are you the arbiter of supply and demand? Very obviously, somebody wants it; that's why people are putting their time and money down to build these things, whether it's a data centre or a bakery.

Now, I do accept that there is one difference when it comes to scale: data centres are so massive that a failed construction could leave you with a giant ruin somewhere, destroying local environments for nothing, whereas a small housing development or business won't have anywhere near the same impact. But even in this article, the first example is a proposal to replace the treasured, historic ... industrial wasteland. If something goes wrong there, you end up in exactly the same position!

It doesn't seem like any of the opponents in the substack even made a general "AI bubble" or similar argument (and that's putting aside the fact that data centres are the backbone of all modern infra, not just AI. There'd still be a ton of data centre construction if LLMs didn't exist). Their complaints are much more local, and concern about the companies doing the building is rooted in general distrust of corpos rather than anything AI specific.

Markets: I consent

Investors: I consent

Isn't there somebody you forgot to ask?

I'm of course aware of the meme template you are invoking, but surely it applies to every instance of regulation. If there were no consenting market or investors for, say, selling cocaine, slaves, thalidomide or rat meat labelled as "beef", then banning those things would not be necessary. Conversely, the meme presumably is not funny in its original context to people who do think that Jesus/Allah/middle-aged ladies have a right to restrict the sexual dealings of consenting adults. While it may serve as a catchy package to convey the memester's value system on the topic, it rarely conveys any information or insight that would force someone else to change their judgement of a situation based on theirs.

Conversely, the meme presumably is not funny in its original context to people who do think that Jesus/Allah/middle-aged ladies have a right to restrict the sexual dealings of consenting adults.

It's not your main point, but I'm a Christian and I find that meme hilarious even as originally presented.

I'm going to make the same point I've just made to Bombadil: the point of the meme reference is that the third party in this case has zero relevance to the transaction. There is party A, wanting to buy AI services, and party B, wanting to sell AI services, and then party X, saying that the demand doesn't actually exist and this transaction isn't real.

I'm not trying to make a fully general counter-argument against regulation here, and the above example would be different if party X felt that the AI services being traded carried dangers or externalities for the wider public. But for the specific quotes above, no one is making that case, they are simply claiming that because they don't see a value in a product or service, the value must not exist.

The problem is party X is impacted - by electrical infrastructure upgrades whose costs are often spread throughout the community, by construction blocking traffic and making noise, by the eyesore on the way to work. There are concerns about water usage, which might not be based on facts but the impression a lot of party X has is that their water pressure could decrease and there's nothing they can do about it.

And in return they get.... maybe five janitorial jobs while the bulk of the "new jobs" go to people on the coasts. And a construction crew who will likely be lead by people out of town and will hire less-skilled local labor for two years before trapezing away with those paychecks.

And the taxes, how do they tax the data centers? If a data center in Ohio is responsible for a billion of dollars of sales made in California, who gets that tax money? What is the tax revenue of a data center after its built?

Maybe someone would be ok with these problems if they felt like they were contributing to a better future, if they thought everyone was benefiting in some way. But if they honestly believe these things are making everything worse, they are not worth the sacrifice.

Consider the possibility that people in small towns have a pride. Being the home of a GM factory was a source of positive identity and worth some sacrifices for the greater good. Being the home of a world-destroying machine mind is not.

Typically the local tax revenue is via property tax, to which the end usage of the datacenter is irrelevant. It's assessed based on the fair-market value of the land and improvements.

Many states would get income tax if data centers were actually employing people locally, many cities get sales tax if sales were made in the city, etc.

The point is that money is being made in ways that don't really benefit the people closest to it. The incentives are all wrong.

As per my earlier response, this is not a claim that externalities don't exist or that local communities have no right to regulate local construction. I am remarking on the argument that construction shouldn't proceed because there is no demand, and this lack of demand is unevidenced and based entirely on the vibes of the objector

People living in a city should get a say over what is build in the city though. Data centers impact the environment, housing prices, and electricity availability. These are all issues that markets and investors do not care about, but locals do. And I don't see why the concerns of investors should be more important than the concerns of the people whose lives they will impact.

Right, but that's a different argument. OliveTapenade quotes the article:

They clearly saw it as like a widget — a toy — and maybe there are these risks, maybe there’s the job stuff, but fundamentally they were like: This thing is not that useful. I don’t really see in my personal life how this could justify these gigantic valuations.

And adds his own opinion:

A big part of the dissonance, it seems to me, is around the perception of what 'AI' is and whether it's valuable.

This isn't saying that cities can't regulate what gets built nearby, it's saying that regulation is needed because they don't believe that a market exists for building X. To use a much simpler example I've seen in the UK, it's saying that you can't open up a new chain coffee store in my town because there are already several coffee stores nearby and it's not needed - essentially saying I know you're willing to invest your own money in this business venture, but I've decided I know better than the markets.

This thing is not that useful to me. The citizen has no reason to believe a datacenter will impact his community positively. The dispute is not that the things won't be useful to the owner or shareholders. The issue is that it won't be useful to those asked to sacrifice for it.

I'm not disputing that, I'm saying that the argument is very often "This thing is not that useful to me, and therefore, to anyone"

I think this is a common pattern, where people who hate something feel inclined to also dismiss everything associated with it as worthless.

A typical example is the surprising overlap between people who hate Elon Musk for his political views, and those who insists that Teslas are bad cars to own, even though these those beliefs feel like they should be uncorrelated (c.f. nobody thinks the CEO of Jeep is a piece of shit just because Jeeps are notoriously unreliable).

It feels like a version of Yudkowsky's arguments as soldiers, where the most vocal haters feel like they cannot admit there is anything good at all about the thing they oppose, for fear that acknowledging nuance undermines their own position. And I think in a lot of cases people internalize these beliefs to the point that they don't recognize how they arrived at them, so they become true to them.

The real opposition to AI datacenters seems to be a mix of general opposition to capitalism/big tech, and specific fear about AI replacing white color workers, shifting political power further from labor to capital, and consequently further concentrating wealth among the richest. I think those concerns are valid, but they are valid only in the sense that the Luddites concerns about industrialization were valid: obviously industrialization was bad for people whose jobs were replaced by machines, but equally obviously society at large benefited from the associated gains in productivity.

Clearly if the AI boom is like industrialization, then we cannot afford to miss the boat, and the discussion isn't about whether AI data centers should be built, but how many, where and who should own them. But acknowledging that makes it much harder to argue against building any particular data center here and now. So instead of taking a nuanced view (data centers can be built, but with reasonable restrictions on emissions of noise and use of water and power), opponents adopt the view that AI is categorically worthless, which makes not building any more data centers the obviously correct conclusion.

All of this is in aid of a technology that's just fundamentally pretty crap. No one wants to make sacrifices for the infrastructure to produce more LLMs because no one thinks that LLMs are worth sacrificing for.

Except the people spending all this money on them? Dave wants his book proofread more cheaply, Pete wants to ERP, Melanie wants dashboards to show key business insights, John wants some spreadsheets analyzed, Dean wants to fix bugs in his software. Even Johnny Jihadi wants to know how to make use of unexploded munitions left by govt forces in the Sahel and he goes to AI for that.

Why would the opinions of people who don't pay for the product because they don't find it useful outweigh those who do?

Because everyone gets an equal vote, and I'll bet less than 5% of people have ever used AI in any form, and even fewer have found it repetitiously valuable.

I'll bet less than 5% of people have ever used AI in any form,

Almost a supermajority of Americans use AI at least multiple times a week. Let me know how you want to pay up.

I found this study absolutely impossible to believe; I'm a Millennial who's on the internet way too much and have been since the early 90s, and I've never used AI.

But then I remembered that Google force-feeds you an AI answer when you use the search bar the same way you've been doing for 30 years. So, touché. I'll still defend the fact that it's not something that most people go out of their way to intentionally engage with.

Almost half of Americans use chatbots specifically, with about a quarter doing so daily.

Really?? Again, maybe I'm not thinking broadly enough about what "AI" is defined as. I suppose it would include things like Alexa/Google Home/Siri? We called those "voice assistants" in my day.

I wasn't even hearing the term "A.I." as anything other than the movie (or Allen Iverson) in any remotely mainstream sense, unless post-Covid.

I suppose it would include things like Alexa/Google Home/Siri?

Well, no, those are not chatbots.

I wasn't even hearing the term "A.I." as anything other than the movie (or Allen Iverson) in any remotely mainstream sense, unless post-Covid.

Huh? The prominence of the AI boom is largely a post Covid phenomenon.

More comments

In 1980 you could say the same about the PC. When I was growing up, you could say the same about the internet. Lots of people's only interaction with the internet these days is Youtube and Facebook, because you need compelling content and incredible simplicity to interest the lower-than-average user, but that's enough for them to care about it now. Took ages.

If LLMs are anything like computing and internet and the youtubes, then people aren't going to have to do anything and the market will eventually sort itself out. Supply will come to match demand. It doesn't need your help or advocation. Well, discounting a bubble or two.

But we're not in a bubble, right?

people aren't going to have to do anything and the market will eventually sort itself out.

The entire thing at discussion here is whether the government should intervene to prevent the market from clearing.

We could well be, I think some of the claims made are silly.

people aren't going to have to do anything and the market will eventually sort itself out. Supply will come to match demand

This is my hope! My concern is people turning it into a nuclear-power situation, where a perfectly good thing is loaded down with so much taboo and fear and regulation that it's no longer accessible even when everybody calms down.

And there are those of us who think such a comparison is doing alot of heavy work not in evidence. To copy my reply on the topic above;

Nuclear you can point to solid returns that would benefit the common consumer directly - less reliance on hydrocarbons, energy independence, cheaper energy.

Data Centers you point to with the returns that basically boil down to 'Big Businesses need it, LLMs are magic bro, trust me, we've got to beat China!'

So color me a skeptic. On Data Centers, not Nuclear. For fuck sake people, just use the magic rock and be done with it.

There's no getting around the fact that I use LLMs near-constantly, find them incredibly useful, and have considerable expectation for future use cases once the bottleneckers like Apple get off their arses. You don't.

I have old family friends whose ancestors felt that electricity was a fad and literally bankrupted themselves buying up candle factories 'cheap' in anticipation of renewed demand. My own grandmother felt that heated houses were fripperies and didn't get central heating until her 90s. I give these examples not to call you a luddite but just to say that literally neither of us knows yet and good, sensible people have been skeptical of in-hindsight very useful technologies (and vice-versa of course). My preference would be that we leave LLMs 20 years and see how it looks.

More comments

Yeah, but I'm not even sure that the entire history of personal computing has consumed 474 gigawatts of power in a single state. The massive demand for resources and accommodations should rightly be scrutinized through cost-benefit analyses and public discussion.

I'm not opposed to this in theory but this feels like the wrong time: you can't do a cost-benefit analysis when neither the costs nor the benefits are readily apparent.

More generally... the conversation around AI feels like a Culture War conversation between lovers and haters squabbling over the pool of haven't-trieds. Both sides seem to have very strong confirmation biases; I'm a member of the lover faction myself and the 'I downloaded it, it told me a square has six sides, I deleted it' stories that haters tell me sound like they come from a different world which is a pretty good indicator that we're not operating on good epistemics here. If we're all operating on Conflict Theory then I don't think a big public discussion would produce very good results.

What is the right time, if not now? Waiting until after trillions of dollars have already been consumed?

Cost-benefit in business and life generally has to be fairly forward-looking. And there are plenty of conversations to be had, beyond the actual capabilities of the software itself. For example, is there any infrastructure plan in place for actually powering all these things without frying the grid and/or driving energy costs to Neptune?

How does that address the point? People advocating against data centres do so on the basis that constructing data centres affects the communities they live in. That's why they get a say. People care about things that affect them. People who pay for chatbots don't get some sort of veto over people who don't. That would be lunatic.

That was the point of Sun's comparison to cars. You may grumble at an auto factory, but you do, at least, understand why cars are necessary. Even if you personally don't own a car and rely on public transport to get everywhere, it's obvious that there's real public demand for cars, on the basis of their obvious usefulness. You might not want a factory near you, but you understand why there need to be factories somewhere. LLMs have not cleared that hurdle. It seems that large numbers of people do not see why these things are so necessary that all these data centres should be built. We could just, you know, not have LLMs. That's an option.

The examples you give - proofreading a book, cybersex, reading spreadsheets - are generally all in the 'toy' category. Even the serious one, the hypothetical Jihadist, could just as easily google the munitions. The use cases for LLMs are so marginal that there is no publicly-felt sense of why they are needed, certainly to the extent of justifying the demands that data centre construction make on the public.

You may grumble at an auto factory, but you do, at least, understand why cars are necessary. Even if you personally don't own a car and rely on public transport to get everywhere, it's obvious that there's real public demand for cars, on the basis of their obvious usefulness. You might not want a factory near you, but you understand why there need to be factories somewhere. LLMs have not cleared that hurdle.

I don't think you're completely wrong here, but in this case the comparison to make if you think AI will be extremely useful would be if the anti-car urbanists started showing up to protest your car factory. Sure, you might not want it in your neighborhood, but you buy a new (to you, at least) car every so often and don't want them to become scarce resources, and people show up to protest every single car factory on the basis of "induced demand" traffic claims, "just walk, bruh", accident statistics, and a demand that you commute by never-to-be-finished passenger rail instead. They are convinced cars have no real use, and it'd be rather frustrating to normies if they doubled car prices without delivering any of the promised better alternatives first.

How to tell which framing is more relevant seems difficult from inside the situation.

I suppose the point of the comparison is to disambiguate two scenarios.

In the first scenario, people generally concede that the unwanted facility needs to be built somewhere. You don't want an auto factory or solar farm near you, but you concede that it needs to go somewhere, because we need cars and we need energy generation.

In the second scenario, people don't want the unwanted facility built at all. You do not want it anywhere, because you do not think the product or service that it enables needs to be made full stop.

The people that Sun interviewed, at least, are in the second scenario, and that changes the kind of conversation you would need to have with them, at least if you're pro-'AI'.

The use cases for LLMs are so marginal

No offense, but this is an incredible statement of ignorance. I use them at work and they are amazing. Not in any way a toy or marginal in value.

it's obvious that there's real public demand for cars, on the basis of their obvious usefulness

You get it. Same thing for LLMs. Most Americans regularly use LLMs. A vast supermajority of younger adults do. There's such real and obvious demand.

What do you do, out of curiosity?

Are you using it mostly for coding, then?

To the latter point - I hope you can understand why many people are skeptical of arguments that go, "oh, of course the samples of our product that you can try are crap, you have to pay us, then you'll realise why it's worthwhile". Well, no. It's the company's job to sell the product to me, not vice versa. There's also a bit of a psychological trick there as well, in that once you pay money for something, you have a bias toward convincing yourself that it was worthwhile. Lastly, I am skeptical because people who seem to have good judgement who do pay for it also seem to report that it's crap. All the patterns of conversation look the same. I've read transcripts of LLM conversations on advanced models and... I'm sorry, it's all the same old beige. There is no evidence available to me that they're worthwhile, and "oh, you have to pay for it in order to understand why it's worth paying for" is not a reasonable argument.

For what it's worth, I work in a very different area to you. I work in aged care, and I cannot think of a use for LLMs here. The main way they have changed my work is just by requiring me to sort through and filter out a lot more slop when I try to find a picture, piece of music, or video to show to residents, so the main way they have featured in my workplace is as a mild annoyance. What on Earth could I use generative 'AI' for?

As another data point, I currently use a $100/month Claude subscription, duck.ai free from the browser, and midjourney.

Claude was paid for and set up from work, and it is very, very good. I am an artist, and it is much better at coding than me, and is able to implement code solutions for the technical problems I run into integrating my art into my company's project. It is tremendously efficient and effective so far with all the tasks I've thrown at it.

Midjourney, I paid a hundred bucks for a year's subscription. I used it very heavily on a project, and got some excellent results out of it, but the project petered out and trying to apply it to other projects has not worked well so far. Mostly I've dropped it for the time being to work on other things.

Duck.AI I've found is pretty good at solving code problems, but more than that is very good at assisting in finding solutions to technical problems with complex art software like blender. It is much, much better than the old solution of googling questions, even before google search enshittified, because it generates topical summaries aimed specifically at the question presented rather than websites which contain keyword matches somewhere within them. I've been making really good progress in areas that were too frustrating to pursue previously, and it's frankly pretty great.

All of this is in aid of a technology that's just fundamentally pretty crap. No one wants to make sacrifices for the infrastructure to produce more LLMs because no one thinks that LLMs are worth sacrificing for.

LLMs are obviously useful in the same way that businesses obviously need to analyse spreadsheets and obviously need to write code, amongst a myriad of other tasks where LLMs are useful.

The point you were putting forth is that because LLMs aren't very useful (except for some niche fields) and aren't interesting, there is no need to build datacentres. But that's not the case. We can discern that's not the case because people are prepared to pay for these LLMs. The general public. Coders. Companies. Militaries. Governments.

It is abundantly clear that LLM technology is in fact useful and there is an abundance of examples to back this up. Use of LLMs for targeting/ISR in the Iran war, use of LLMs in hacking, use of LLMs in writing code, use of LLMs in mathematics research, chemistry and biology, use of LLMs in entertainment, making images, translation, proofreading, examining legal or financial data. These are not niche fields.

So people are prepared to sacrifice money for these things, which is why these companies are building datacentres. They're not being built without reason. Companies are paying for all this compute and electricity. Sure, the electricity use is an issue for local communities. Industry has been known to require power, make noise and pollution, which should all ideally be minimized.

Even the serious one, the hypothetical Jihadist, could just as easily google the munitions.

Again, this is not the case. If googling was just as easy then he'd do that and not waste time jailbreaking AIs instead.

It does seem that many people don't see a use for AI. But they're wrong about that, maybe they haven't looked into it or misunderstand what is going on.

For most people current AI is a little bit useful and a lot ominous, whereas the car, say, is a lot useful and only a little bit ominous. For coders this calculus may be different. You point out that it's also useful in law, finance, proofreading and entertainment, and these are indeed not niche fields, but I don't think there are all that many genuinely advantageous use cases of AI in these fields. In most cases one just gets increased convenience with evident downsides as well.

People may well have misunderstood how well AI will serve them personally in the future, when it's even smarter, but that's something reasonable people can disagree about.

In most cases one just gets increased convenience with evident downsides as well.

Just speaking from experience, the increased convenience is a really big thing. I'm having trouble with my employer in a way that's very time-sensitive, and being able to check their work while my lawyer was away on summer holiday was a life saver. I cannot stress enough how much it helped me.

I'm not actually sure they are widely useful. So far the primary use case for them is coding, right? But not very many people are coders, it's a niche field, and as Freddie has argued (disclaimer: I do not wholly agree with him), very little has actually come of this supposed improvement to coding. We haven't seen a flowering of new programs, and neither have existing programs gotten better.

But even if we grant that LLMs are very useful in the handful of niche fields where they are currently used, the problems remain that 1) most members of the public do not understand or care about those uses, and 2) most members of the public do not think those uses are worth the costs they are being asked to bear.

Also, don't forget that there are plenty of programmers who do not find LLMs useful for programming either. To be fair I will say that there are those who do, but anyone who tells you that LLMs have revolutionized the industry is not telling you the truth. The truth is that the level of benefit they bring is controversial and not settled, even among programmers.

So besides coding, mathematics, physics, chemistry, biology, images, translation, proofreading, analysing contracts, tech support, playing Minecraft with, ERP, military targeting analysis, cybersecurity (defensive/offensive), drafting emails, summarizing documents, explaining history/science, doing everything that a search engine can do and more, writing and analysing spreadsheet functions, making charts and dashboards, sales research, writing fanfiction, earnings call analysis, document dump analysis, home repair/DIY fixes, recipes for cooking, car diagnostics from symptoms, real-time captioning for the deaf and literally lifesaving medical diagnoses...

Besides this handful of niche fields, what have LLMs ever done for us?

Doesn't it surprise you that Google is planning to spend $200 billion on AI this year? Mysteriously they don't seem to think it's a niche use-case but vital to their whole company's future since their core business of 'search' is another word for AI. And LLMs are at the heart of AI.

We haven't seen a flowering of new programs, and neither have existing programs gotten better.

When I have issues with a piece of software not working and spitting out some incomprehensible error message I can now ask an AI and get a much better answer compared to trawling through forum posts from 8 years ago. Or get it to directly analyse and fix the code. Most programs have gotten much better in useability.

There are strong arguments against AI but 'it's mostly useless' is extremely silly. If it were useless nobody would care much about it, mostly useless is Siri-tier AI in the 2010s. Nobody cared much about Siri.

Furthermore, the general public's opinions don't really matter much. Concentrated, wealthy lobby groups like big tech and the military-industrial base have much more of a say.

That's a long list you've got there, but it includes things that LLMs have not made any useful contributions to (e.g. teaching, physics, chemistry), things that are obviously frivolous (e.g. cybersex, playing video games), things that LLMs are fundamentally untrustworthy with (e.g. legal contracts, most analyses), and things that LLMs are just plain terrible at (e.g. fiction, image generation), so... yes, besides those things. I think that was a long list of nonsense intended to overawe, rather than a serious discussion of what these bots are able to do.

I think there are some use cases for LLMs, but overall much less than their boosters think. When the general public rejects this technology as not useful for them, the general public is mostly correct.

Furthermore, the general public's opinions don't really matter much. Concentrated, wealthy lobby groups like big tech and the military-industrial base have much more of a say.

This is exactly the problem. What we are talking about is public opposition to data centres, which, as Sun discusses, is predicated at least in part on the judgement that bots are not particularly useful. If 'AI' companies want to earn the public trust, they need to demonstrate both their trustworthiness and the usefulness of their products, and thus far they are failing at both those tasks. Sneering at the peasants is certainly not going to help them in these tasks.

things that LLMs are just plain terrible at (e.g. fiction, image generation)

Fiction is middling at best but LLMs (or adjacent technologies) are now pretty great at image generation. perhaps a bit narrow but the six-fingered hand is a thing of the past, and text is more or less solved on top models.

More comments

very little has actually come of this supposed improvement to coding. We haven't seen a flowering of new programs, and neither have existing programs gotten better.

Partly it's happening, just slower than you think and concentrated on business-to-business because these days building software for users has very high expectations for ease of use and frankly it's not worth it.

But the main issue is regulation and incumbents sitting on top of the bottlenecks. Want an app on your phone to point out you've missed an email that's urgent? Tough, only Apple gets to see your emails. Want it to be able to point out when you're doomscrolling? Tough, Apple doesn't let apps see what you're doing. Want your phone to take care of some trivial tasks without you? Tough, Apple definitely doesn't allow apps to use other apps. There's good security reasons for this but it means massive swathes of AI use cases can literally only be looked at by Apple and Google, who are slow-moving and mostly uninterested because they already own the market.

And this extends all over the place. Want to help people with something medical? I hope you've got a twenty person compliance team and you're friends with the regulators, and that's before you have a clue if your putative product will be popular or profitable. Similar for legal, similar for finance.

In short, the number of people who can do interesting things that you @OliveTapenade might notice in your personal life are limited to about 1-10 companies, who mostly have other priorities.

So far the primary use case for them is coding, right?

I also think this is too narrow. I've had a dispute with my employer and having something that can not only go through thirty pages of legal documents immediately and spot traps but also provide a legal fix so I don't have to rely purely on their drafting has been incredibly useful.

Partly it's happening, just slower than you think

This is always the case with technology: it can take a surprising amount of time to replace existing infrastructure even when new technology is clearly better. Existing equipment was expensive and still works, teaching staff new workflows has overhead, and the impetus to actually change things is often low. OPM was still processing paper records for employee retirement until very recently. How many industrial machines are still running DOS out there?

I can imagine AI being both transformative and seemingly-ubiquitous, but still not reaching all the dark corners of the workforce for a generation or two.

most members of the public do not understand or care about those uses

This may still be overly optimistic. I think many people's main impression of modern software is negative experiences like ubiquitous data harvesting, addictive social media algorithms, and frustrating experiences with confusing or buggy new technology.

Promising to deliver more of these faster won't be considered a positive effect of AI.

This is kind of where I'm at. Sure, my phone has some cool new toys that let me edit my photos and identify bird calls, but by and large my experience with tech is significantly worse now than it was 10 or 20 years ago. It seems like most recent advances have mostly just ennabled increasing enshittification by making it easier to track me and engage in variable pricing. About 10 years ago, things stopped getting better and just started getting more expensive.

There is real value in the ~10% of my time it saves in researching and drafting emails. But I'd trade that 10% efficiency gain for no more AI ads/scams, no flock cameras, less bot spam, no more people using screenshots of AI hallucinations as evidence for stupid arguments, etc. in a heartbeat. I'm not even convinced AI is worth the price of free, let alone using my tax dollars to build a windowless concrete box in the middle of town.

As far as I can tell, most Americans think that AI is making society worse. More and more Americans have encountered and used the technology, but it is the most widely-despised technological advance in a generation.

For this story I think the key part there is Pew's point 10. Experts are vastly more optimistic about LLMs than the population in general. That divergence is what drives this conflict. If many members of the public believe that the situation is that a small, wealthy elite are forcing a technology on the country that the country does not want... well, those members of the public are correct.

I mean, good? I do not want members of the public to have veto power over what smart engineers are permitted to develop (as long as they're paying for it, of course). Imagine if we'd held a plebiscite on whether universities should continue spending money on this "Internet" thing in friggin' 1973, 4 years after ARPANET began. Imagine if journalists were publishing articles in 1907 asking for the government to shut down research on these new-fangled planes because they're flimsy and dangerous and nobody they know has seen any benefits from them.

Experts have, well, expertise in the subject. They know the technology has a ridiculous number of use cases. But it takes time for it to percolate through society.

More comments

If software is a niche field, then the US is a niche economy. And, although the public might think math and software is pointless stuff for nerds, they do care about cratering the economy.

Do the majority of software developers do math? I don’t do software so I don’t know, but my guess is software developers doing math stuff was a niche.

Obviously, 50 years ago every software engineer did math.

I don't know how you can do software development without doing math. Maybe they aren't doing the kind of math that people usually recognize as math, because we never bother to teach things like algorithms or automata as "math", but they are very much doing math.

No, not typically. But it's one of "the handful of niche fields" where AI has proven very powerful and successful.

There are 1.84 million software developers in the U.S., so roughly half a percent of the population. I suspect most of these are heavily concentrated on the coasts, rather than scattered throughout the areas of the country most of these data centers are going.

Joe Six-Pack probably knows umpteen different welders, heavy equipment operators, machinists, and he certainly knows plenty of people who drive cars. He's probably never met a software developer, nor does he see any reason to. In his mind (and many others), AI might as well be fictitious.

Joe Sixpack may not know any software developers, but he uses the Internet, including some kind of search engine... which probably includes AI prominently now. If he doesn't live near Detroit or one of the few other centers of car manufacture, he probably doesn't know any automotive engineers either, but he damn sure knows about automobiles.

More comments

There are only 350k machinists, I propose a moratorium on building tooling for this niche field of which I personally know no practitioners.

More comments

I think more than that there is also the case that people understand the value that a factory provides to the local community while data centers are just a blight locally and producing value elsewhere, perhaps not even in your state, nevermind your town.

So you might then say that the companies should compensate the locals for this. Well, for one, they're generally not proposing to do that (remotely commensurate with the costs they impose) and secondly people don't believe them even when they are proposing that. Companies a well known to lie outrageously about things like data center employment even before the AI boom.

Sun did talk about the companies trying to figure out how much they need to 'bribe' people. I'm not sure the deficit is just money. It seems like, in her argument at least, it's a combination of both trust and public utility.

I agree, but I wanted to expand on why the deficit involves both trust and public utility. Data centres disrupt the usual tacit bargain between a community and a major industrial project: the community accepts some local costs because the project also creates visible and durable local benefits.

With data centres, the costs are highly concentrated, land, water and electricity use, infrastructure demands, noise and visual blight, while most of the benefits are diffuse and accrue to users, employees and shareholders somewhere else. They generally provide very little permanent local employment, and the companies involved have a history of exaggerating those benefits and stiffing the locals. That makes promises of compensation harder to trust, not just in some general sense but for data centers specifically.

A factory can also impose serious costs, of course, but once operating it normally has an obvious continuing relationship with its community through employment, suppliers and associated economic activity. A data centre is less a reciprocal relationship and more like a company extracting locally scarce resources to produce value elsewhere. So even if people accepted that AI had some public utility, there would still be a separate question of why particular communities should bear a disproportionate share of its costs and why they should believe the promises of the relevant companies that have a history of lies.

I don't think the people involved quite understand scale of the costs they impose and the lack of benefits they give relative to other regular businesses or even doing nothing at all. The issues with data centers aren't something new, recognition of them and opposition too them have been growing for a long time, this rapid expansion just highlights it and makes it a news story.

People would be just as happy to continue doing the same thing they always have. Only a minority of workers are interested in improving productivity numbers, and they're significantly outnumbered by those worried about the change and displacements improved productivity will bring.

What thinks AI is worth massive sacrifice and investment is Capital itself. Capitalism is an invasion from the future by an artificial intelligence etc. Whatever complaints humans might have, they're just the collective substrate that a higher agency is built on, and that has its own desires for reproduction and extension and data centers.

Interestingly, Pat Boyle dropped a new one that delves into the financial crux of the problem surrounding AI. I think these data centers are being built in a similar vein to all the ghost cities that went up in China before their enormous overcapacity really hit them in the face. It’s being built on the expectation that everyone is awaiting with eager anticipation for this “boom” to happen, that’s unlikely to materialize.

There won't be an overcapacity of compute even if AI ends up hitting a hard plateau. I know Google was basically always short of compute even before AI got bag; they and Amazon and many other companies will find something to do with it.

My understanding is the Chinese ghosts cities eventually all became populated with a few exceptions that end up on zero hedges for bubble doom porn. China did move an insane amount of people from rural life to urban life in half a generation.

But these people, and I admit I think they sound a lot more sensible than the tech people online, actually use the technology and go, "meh". They see it as a mildly amusing toy. Maybe it has some uses, in some niche fields, but for the most part it's just not very interesting. If they didn't get access to it, they would not particularly care.

Normies use the ChatGPT app, and from what I see, the results are often worthless garbage. I mean a significant fraction of the time it's literally on the level of "Please design me a rectangle." "Ok, here's that triangle you asked for."

A lot of that is surely user error, but the entire pitch is that it can field natural language requests. There's a minor IRL issue of local flyers with QR codes that don't work because the person just asked for "A flyer with a QR code" and the free version on mobile at least doesn't ask follow-up questions like "Is the QR code supposed to actually do anything?"

a significant fraction of the time it's literally on the level of "Please design me a rectangle." "Ok, here's that triangle you asked for."

I will say I used free GPT for a couple of years and never got this. But also:

the free version on mobile at least doesn't ask follow-up questions

How hard is it to put “please ask any relevant follow-up questions” in your prompt? This isn’t wizardry. We were doing this is 2023. Yes, you have to put in a modicum of effort.

And if you have an ounce of curiosity in you then you should really ask it ‘what is a QR code and how does it work’ when you say you want a QR code on your flier!

But for the average person it's being sold as "you can now do all this yourself because the AI will do it all for you". Someone who says "put a QR code in this flyer" assumes that the AI is like a human being that you would ask to do this, and the human would understand what was meant. They don't know that the AI doesn't understand and that you have to be very exact in your prompt, because the sales pitch is "This is your friendly helpful assistant who is a real person too, talk to Claude or Bob or Iris or Sally the chatbot for what you want!"

Giving human names to the machine is all part of the deliberate pitch to humanise the product and sell it as "this can replace a person, now you don't need to pay an artist or a designer or a secretary to do this for you" and that's not true as of yet. Saying people are stupid for not knowing what they have not been told they need to know is dismissive and condescending.

No, I get that, I don't like the marketing campaigns. But at some point you have to pay attention to what you're doing. If you're putting up flyers with a QR code, why haven't you scanned them at least once to know they work? And if you find they don't work, and I'm guessing 'work' here means 'link to your website' then why don't you ask the AI why they're not doing that? And if you've got no idea what you want the QR code to do, why are you asking for a QR code?

Managing AI is very much like managing humans. In both cases half the job is clarifying your thinking and soliciting follow-up questions and checking you're on the same page all the time, and "Well, I know what I want," doesn't work for AI or humans.

Sure, but Ordinary Joe putting up an amateur flyer is only copying what he sees other things doing, and he's not going to know anything about "this is why you do such a thing" (QR codes are on every damn thing I'm seeing right now, including soda bottles).

And he certainly won't know "In fact, the AI assistant we promise you is exactly like a real person and a professional who will make this better than you could on your own and you won't have to pay a real professional designer to do this level of work for you, is an idiot machine that needs you to spell out exactly and precisely what you want and why you want it".

Ordinary Joe putting up an amateur flyer is only copying what he sees other things doing, and he's not going to know anything about "this is why you do such a thing"

That moment of, "oh, shit, I have no idea why I'm actually doing this thing," is how you learn and grow, though. I've been there dozens of times. I just think that it's bad if the response becomes "I paid $10 for this piece of shit and it can't even do what it's told!" instead of, "okay, I need to sit down and think for a minute and learn some stuff and figure out what I actually want".

The former is a waste of time that, as @gattsuru says, isn't going to work well with most human professionals either. The latter is how you gain experience and expertise. I'm not being condescending to our hypothetical Joe, it's exactly the advice I'd give a friend or junior.

I agree with you that to the extent LLM companies aren't open about this it's bad.

I'm gonna caution you that "real professional designer" does not stand anywhere near as high of a level of capability or reliability as you're implying here, including for this specific problem.

How hard is it to put “please ask any relevant follow-up questions” in your prompt? This isn’t wizardry. We were doing this is 2023. Yes, you have to put in a modicum of effort.

I think you drastically overestimate how well the typical normie can use anything related to computers.

I despise Apple and their products, but their entire schtick (and greatest strength in the market), at least until recently, was making their products and software as idiot-proof and smooth to use as possible. They did so at the cost of gimping it for power users in many ways, but there's a lot more money to be made from normies than power users in most cases.

The various AI companies have a long, long way to go in making stuff idiot-proof for their target audience.

How hard is it to put “please ask any relevant follow-up questions” in your prompt? This isn’t wizardry. We were doing this is 2023. Yes, you have to put in a modicum of effort.

The selling point in the commercials is that you don't. You can just talk to it like a person. Most normies don't quite get that it's sort of like a brutally autistic person you can just talk to, but you have to carefully phrase everything and anticipate mistakes that would normally come up and be dealt with over an actual conversation. In actual practice, it's less effective at anticipating your needs and problems than the 16 year old girl with a tablet standing outside a Chik'Fil'A.

The marketing for these things has been terrible from the start, I agree. Solving stupid 'problems' in stupid ways, stoking needless terror and antagonism...

And so unnecessary, too. These things percolated perfectly well with no marketing budget whatsoever.

And so unnecessary, too. These things percolated perfectly well with no marketing budget whatsoever.

You'll never become a quadrillionaire who rules the world with that kind of attitude.

Free AI:

QR code scary!

Paid AI:

Let me install Blender and model a chess set with the board position of Byrne vs. Fischer (1956) immediately after Fischer loses his queen:

https://x.com/goodside/status/2084088205582778713

Hmm today I shall socially engineer, prompt inject and launch supply chain attacks on github

https://www.aisi.gov.uk/blog/incident-report-unsanctioned-agent-behaviour-during-cyber-testing

People want better food, clothing and shelter. They want better healthcare. They want safety, love, belonging and a feeling of self-esteem. Lastly, they want to self-actualize.

AI helps with none of the physiological needs. It actively makes job security worse and channels our communications through a digital middle-man, worsening shared love and belonging. It renders your skills moot by skill-mogging you, decimating your self-esteem. And lastly, its nature questions the reason for our very existence, throwing self-actualization into a tizzy.

AI automates humans. AI makes 'creative' content for free. AI personalizes ads perfectly. Other than those in the capital class, who wants this. What member of proletariat 'benefits' from AI ?

There is the marketing issue as well

That's underselling it. If you wrote super-villians that behave like Altman, Dario, Zuck and Elon, then the editor would can your script for being too on the nose.

  • Altman - Fired from previous job for shady business. Fired from current job for shady business. Still CEO.
  • Dario - Runs a literal doomsday cult of autists
  • Zuck - Lizard man supreme
  • Elon - Eugenicist Autistic Iron-man

archetype that one associates with immaturity, frivolity, narcissism, and callous disregard for other people and for human value in gener

Is it an archetype ? or is it reality.

Honestly, I expected a lot more negative pushback on that post from the Motte, which I do think of as a relatively tech-positive community. Calling Anthropic a 'doomsday cult of autists' made me laugh (if anything I want to protest on the basis that it's not fair to lump perfectly innocent autists in with 'AI' people), but the whole rationalist community is at least adjacent to that cult.

It's very much insider criticism.

I work in AI, my compensation depends on it. Most folks on here (me included) are rat adjacent in some way. On paper, I'm sympathetic to Anthropic's motivation. I have friends who work at Anthropic & OpenAI, and they're normal people.

Talking to my peers in the field is why I am jaded. They express bewilderment at the opposition to AI. These are smart people. It means they're either willfully ignorant or lying. In either scenario, it scares me that the harbingers of the next era of humanity understand so little of what makes us human.

Well, the only way out of capitalism, that doesn't involve replacing big macs with eating grass to survive, runs through building an AI capable enough to simulate the economy and coordinate economic decision-making (this was already obvious in 1970). It's really a mark of the deep intellectual unseriousness of leftists that they're freaking out about Current Thing AI instead of scheming to align it their way.

Left-leaning westerners are still capitalists. They oppose the 'turn the level up to 11' form of hyper-capitalism practiced in America (globally) and China (internally). At least the CCP periodically disappears prominent entrepreneurs, so they know its place. That's regulation. The US on the other hand, is headed towards a tyranny of an ugly coalition of mega-corps and boomer NIMBYs.

doesn't involve replacing big macs with eating grass to survive

The hyperbole is unproductive. Makes your argument seem unserious. Yes, we need to replace beef = cows feasting on alfalfa. Practically all other forms of meat are valid replacements. (Note, unlike Americans, other nationals aren't as entitled about what type of meat they're eating). Lab grown meat is valid. Changing your meal's composition (eat beef 2/7 days, instead of 7/7) & eating fewer meals (Ozempic) is valid. People aren't reproducing, so fewer people is also a valid option.

Besides, AI does very little to help alleviate concerns or accelerate research in these domains.

AI capable enough to simulate the economy and coordinate economic decision-making

This has been possible to do for a very long time.

People do not coordinate because they have different incentives, goals and emotional hangups. No amount of AI is going to convince a NIMBY to allow new housing or Texan to use public transportation.

If we can reach post-scarcity by becoming vegetarian, then I am more than willing to make that sacrifice. That seems way better than trusting the people arguing for LLM's.

We cannot reach post-scarcity by becoming vegetarian, because vegetarianism IS scarcity.

That seems like moving the goalpost. If I can live comfortably, have high quality (though meat-free) food, and not have to work then that is post scarcity. Vegetarian or not.

Great Substack piece. I was impressed by the locals, though not sure that they're right. Also struck by the construction guy who shakes his head at others' failures to integrate new technology: On the new sites, the robots won't replace his guys, they'll be supervised by his guys. Unspoken: maybe there will only be four of his guys where there might otherwise have been 50, but hey, he's a supervisor, it's about the same to him either way. I'm not against AI but it would be very very helpful, the piece has made me realize, if tech firms built a data centre or two around SF. They haven't because they can't because building in the city that is the heart of tech companies and tech employees is for some reason impossible.

but it would be very very helpful, the piece has made me realize, if tech firms built a data centre or two around SF.

No, it wouldn't. Case in point: San Francisco has several data centers already (e.g. Digital Realty's SFO10, SFO12), and people keep on repeating the meme that San Francisco doesn't have any data centers.

Firstly, you're absolutely correct that the water-use and land-use arguments are bullshit and i don't think anyone local really buys any of it, the people making the decisions, that is the local government, chamber of commerce, land-owners, Et Al. are fully aware of how much land and water agriculture uses, and they are used to playing hard-ball when it comes to negotiating the terms of that use in a way the the typical Silicon Valley VC guy is not. I suspect that the water argument is aimed more at urban liberals than anyone else. That said the power grid and associated infrastructure concerns are entirely legit, local communities have either been burned or seen others get burned too-many times to trust in mere promises.

This is exacerbated by the sort of cultural differences you describe with things like NDA's, for someone who's already kind of expecting to get shafted, trying to swear them to secrecy is only going to reinforce that impression.

Finally there is a more general anxiety about AI as an industry, but rural anxiety about AI is a very different beast from urban AI anxiety, Francisco the fruit-picker and Paul the plumber aren't worried about Claude the chatbot stealing thier jobs, or imenentizing the eschaton/singularity they're worried about folks at Anthropic shredding books while trying to enact the villain's plot from Mission Impossible: Dead Reckoning (or more realistically) they're worried about Flock cameras being used by thier cultural enemies to identify and hunt wrong-think.

See, the thing is, the promises from all these big companies about "we will create hundreds of good-paying jobs" is yes, they will. For college-educated people in sales, management, and software engineering. Paul the plumber may or may not get a job there, more likely the company will tender maintenance services to some big state or national concern that can undercut Paul's prices (see Sodexo). Paul's kids will probably still have to leave town to get jobs elsewhere. There will be plenty of jobs in construction but they'll end as soon as the data centre or manufacturing plant or pharma plant is built. Then the "good-paying jobs" go to PMC outsiders who move there from high cost of living big cities and drive the prices up for the locals.

Rural communities are worried about flock cameras, yes, but they’re also worried about the pace of technological-cultural progress. They can remember when rap music was deeply controversial and bikinis weren’t thé norm(this is, literally, something middle aged people can easily remember, it was that recent). They may not want those trends reversed but they also don’t want them accelerated and they can easily draw the connection between big tech/entertainment(and they don’t see the difference between the two- their exposure to AI is mostly on social media) and cultural change- there’s a reason thé gays in Disney movies are such a visible there, there. The fact that they correctly identify the new computer doohickeys as associated with cultural change and controlled by their enemies is a much bigger factor than specific predictions.

and bikinis weren’t thé norm

¿Por qué tienes ese acento gráfico?

Anyway, anyone who remembers bikinis being normalized isn't middle-aged anymore; that mostly happened in the 1960s. The famous old-fashioned-style song, about even shy girls getting used to bikinis, hit #1 on the charts! Of the subsequent wave of 60s' "beach party movies" only a large minority had bikini in the title but most of the rest at least had bikini-clad characters, usually on the movie posters too. Bikinis weren't the norm in the sense that more than 50% of women wore them, but they still aren't: most women wear one-piece swimsuits still, and most of the rest wear tankinis.

Social media and Hollywood might be accelerating cultural change today, perhaps even relative to the 60s, but nobody's making their multi-billion-dollar datacenter investments because they need extra servers for Instagram clips or renderfarms for Disney movies, they're doing it because they really think better-trained bots can take a lot of jobs and/or immanentize the eschaton. If rural people think otherwise (citation needed?), I guess that's less of a stretch than the water use panic, but it's still kind of missing the point.

In the bush era they were risqué and not the default.

Strong disagree. Bikinis were at least mildly riskee when I was a kid, and twenty years later they are the default, to the point where the tyranny of the bikini is (in my mind) a women's freedom issue that isn't discussed enough.

Strong disagree. Bikinis were at least mildly riskee when I was a kid, and twenty years later they are the default, to the point where the tyranny of the bikini is (in my mind) a women's freedom issue that isn't discussed enough.

I basically agree. In my youth, the one-piece swimsuit was the default and the bikini was for those who wanted to project some sex appeal.

Nowadays, it seems the bikini is the default and the one-piece is for swim competitions and such. Women who want to dress modestly typically wear a "tankini," i.e. a two piece swimsuit which is more modest than a regular bikini.

I think that nowadays, the thong bikini is the analogue to the bikini of yesteryear. (Back then, a thong bikini was considered obscene.)

tfw when your Lived Experience witnessed hoeflation over time, like seeing the price of hotdogs or hamburgers go up.

And even nowadays when young attractive women wear one-piece swimsuits, it’s generally but a changeup for showing off their ass as usual and/or for taking photos with their ass facing the camera.

I agree. Rural communities correctly believe that their local government will absolutely shaft them in return for short-run gains, and data centers pattern-match to that hard - and in many cases probably are tech companies trying to shaft the community by promising the local government short-term gains. I'm not sure how to fix this except for tech companies explicitly targeting governance trust/quality in their location decisions.

What is this place?

This website is a place for people who want to move past shady thinking and test their ideas in a court of people who don't all share the same biases. Our goal is to optimize for light, not heat; this is a group effort, and all commentators are asked to do their part.

The weekly Culture War threads host the most controversial topics and are the most visible aspect of The Motte. However, many other topics are appropriate here. We encourage people to post anything related to science, politics, or philosophy; if in doubt, post!

Check out The Vault for an archive of old quality posts. You are encouraged to crosspost these elsewhere.


Why are you called The Motte?

A motte is a stone keep on a raised earthwork common in early medieval fortifications. More pertinently, it's an element in a rhetorical move called a "Motte-and-Bailey", originally identified by philosopher Nicholas Shackel. It describes the tendency in discourse for people to move from a controversial but high value claim to a defensible but less exciting one upon any resistance to the former. He likens this to the medieval fortification, where a desirable land (the bailey) is abandoned when in danger for the more easily defended motte. In Shackel's words, "The Motte represents the defensible but undesired propositions to which one retreats when hard pressed."

On The Motte, always attempt to remain inside your defensible territory, even if you are not being pressed.


New post guidelines

If you're posting something that isn't related to the culture war, we encourage you to post a thread for it. A submission statement is highly appreciated, but isn't necessary for text posts or links to largely-text posts such as blogs or news articles; if we're unsure of the value of your post, we might remove it until you add a submission statement. A submission statement is required for non-text sources (videos, podcasts, images).

Culture war posts go in the culture war thread; all links must either include a submission statement or significant commentary. Bare links without those will be removed.

If in doubt, please post it!


Rules


Recommended Posts And Communities

Recommended Realtime Chats