My next book is The Reverse Centaur's Guide to Life After AI, out next month. Pre-order it now, including as a DRM-free audiobook or ebook, at my Kickstarter, and help me continue to prove that DRM-free isn't just the right way to reach an audience, it's also the best way to reach them.
One of the surprise breakout software products of the early web was Lotus Notes, a kind of primitive precursor to all-in-one office productivity suites like GDocs, Office365, etc. It was so important that its creator, Ray Ozzie, was promoted to Microsoft's Chief Software Architect, succeeding Bill Gates himself:
People who remember Notes tend to deride it for its clunky user interface and demi-functional administrative tools. But what made Notes so central to Microsoft wasn't its polish – it was the fact that Notes represented a brokered peace between IT managers, who wanted mainframe-like control over everything their users could do with business equipment, and the users themselves – workers who kept smuggling internet-based tools into the enterprise network on the very sensible grounds that they had a job to do, and these were the best tools to do it.
The arrival of internet-based tools – especially ones that ran in browsers – represented a major challenge to IT departments, who had been long accustomed to dictating terms to their users. If the IT manager and the compliance department decided that the best way to manage disclosure and leak risks was to block all email attachments for outside users, then that was that: no one could send those attachments.
But after the internet arrived on the corporate desktop, employees who needed to get documents to supply chain partners and customers could treat these IT policies as damage and route around them. Just fire up your Hotmail or Yahoo mail window, or hop on MSN Messenger or ICQ or AIM, or drop the file on an anonymous FTP server and send the link to your counterparty. Job done!
IT managers hated this, and to be fair to them, they weren't (always) wrong. These outside tools came from a variety of untrustworthy sources, including malicious sites that pushed virus-infected versions to their users. Also, by evading firewall rules with these tools, users made it impossible to achieve the compliance goals that IT had been charged with enforcing, and it was IT's asses on the line if the company got in trouble as a result.
Foundationally, IT was being asked to do two irreconcilable things: they were supposed to be enabling workers to get their jobs done, and they were supposed to be stopping those workers from doing things that could harm the business. This can't be done, because the only way to eliminate the possibility that a worker will take an action that harms the business is to gag that worker and lock them in a dungeon. Workers need flexibility and freedom to achieve business goals, and that flexibility and freedom means that those workers might (deliberately or accidentally) thwart the business's goals.
What's more, workers will always run into situations that were not anticipated by policy, and if they are denied any agency or initiative, they will fail to get their jobs done. In work, the exception is the rule, hence the importance of "process knowledge" (all the implicit knowledge shared among workers across the firm and its suppliers and customers, which cannot be captured or recorded):
Indeed, there's a form of labor action called a "work to rule," in which workers only do the things dictated by their rulebooks, without taking any of the routine additional measures dictated by process knowledge. Merely by following every rule to the letter, workers can grind a shop to a halt:
https://en.wikipedia.org/wiki/Work-to-rule
Since the dawn of personal computers, workers and IT departments have come into conflict, as workers literally smuggled technology into the business that could do things the IT department had (often arbitrarily and capriciously) prohibited. When Visicalc emerged as the killer app for the Apple ][+, workers snuck these computers into work and used them to sort spreadsheets in ways that IT had declined to permit. They didn't do this to cheat or steal from the company – the whole point was to do a better job.
So it was with the early web: workers discovered a myriad of new capabilities in the free-to-use world of web-based tools and realized how these tools would make them much more effective at their jobs. The fact that IT wouldn't let them do these things was just more evidence that IT – and the managers who set IT's agenda – didn't understand the business as well as workers.
It didn't help that IT managers' first line of defense was the high-tech version of abstinence-only education: "You only think you need your work computers to do this, but really, you don't, so stop trying":
Abstinence-only education never works, but where "you only think you need this" failed, Lotus Notes succeeded. Lotus Notes provided a whole suite of tools that largely (if imperfectly) replaced the universe of free tools that workers were using to evade their IT departments' edicts, so they could get their jobs done. At the same time, Lotus Notes provided a set of management tools that let IT fine-tune how these tools worked, giving them (some) of the controls they needed to achieve their compliance goals.
Like all brokered peace settlements, Lotus Notes left both sides feeling like they'd made a compromise they could live with, giving up some of their goals, but keeping the things that really mattered to them.
It's impossible to overstate how important Lotus Notes and similar products were, because workers demanded the right to use the web on their work computers, and they made those demands so forcefully that managers had to completely re-do their IT policies, lest those workers treat them as damage and route around them. Back then, the tech press was full of stories about these conflicts, as workers insisted that the new technology that was sweeping the nation was so foundational and transformative that they had to be allowed to use it.
What we never saw back then were stories about how managers had to monitor workers to ensure that they were using the web as much as possible. No one had to force workers to find ways to integrate the web into their workflows.
In other words, the story of the web at work was the opposite of the story of AI at work. Today, you can't turn around without reading a story about bosses who are threatening to fire workers if they don't increase their AI usage:
It's conceivable that over the past quarter-century, bosses have become technophiles while workers have fallen prey to superstitious technophobia, but it hardly seems likely. Historically, workers have always been enthusiastic about tools that let them do a better job – indeed, it's a truism that labor-led automation produces improvements in quality, while capital-driven automation increases throughput (often at the expense of quality).
Workers aren't the only typical early adopters who find AI lacking. As a group, teenagers and young adults hate AI:
That's not what it was like during the early web days. Back then, young people entering the workforce were passionate devotees of the web, to the point where the business press routinely ran articles asking how today's workplaces were going to adapt to the demands of these webbed-up workers.
AI boosters insist that the deficits we see in AI – its lack of profitability, its primitive and error-riddled outputs – are no different from the shakedown problems of the early web (and we know how the web turned out!). But this is a profoundly flawed comparison: the early web and AI are very different from one another.
For one thing, the early web may have lost money, but it had great unit economics. Every new web user brought the web closer to profitability, as did every new use of the web, and every new generation of web technology. By contrast, AI has – in the memorable phrasing of Ed Zitron – "dogshit unit economics." Every new AI user makes AI less profitable, as does every new use for AI, and each generation of AI loses more money than the last. AI is the money-losingest endeavor in human history:
In other words, the early web was a technology that grew more profitable every day, which workers and young people had to force on their bosses – and AI is a technology that grows less profitable every day, and bosses have to force it on workers and young people.
Now, it's true that some workers don't have to be forced to use AI. Workers who enjoy a high degree of autonomy (that is to say, workers who are positioned to ignore workplace coercion) can adopt AI in ways that they feel suited to, just as those early web users and Visicalc smugglers did. They can fulfill the maxim that labor-driven automation improves quality, while resisting capital's insistence that automation be used to increase throughput at quality's expense.
They can act as centaurs (workers assisted by technology), not as reverse-centaurs (workers who are recruited to serve as peripherals for machines). As with all technology questions, what the technology does is nowhere near as important as who the tech does it for and who the tech does it to:
And there's another group of workers who adopt AI voluntarily: workers who see that AI can do a lot of work that they view as dull and unimportant for them. These workers might be right – there are plenty of bullshit jobs out there:
But it's also possible that they're wrong, and they're substituting AI for something that really should be done by a person.
But on the plus side, at least no one has to force them to adopt AI.
If you'd like an essay-formatted version of this post to read or share, here's a link to it on pluralistic.net, my surveillance-free, ad-free, tracker-free blog:
I'm on a tour with my new book Enshittification: catch me next in Los Angeles, Calgary and San Francisco! Full schedule here.
When the AI bubble pops, what will remain? Cheap GPUs at firesale prices, skilled applied statisticians looking for work, and open source models that already do impressive things, but will grow far more impressive after being optimized:
The AI bubble companies are scams. They've spend most of a trillion dollars in capital expenditures, and by their own (very cooked and dishonest) numbers, they are grossing a total of $45b/year, industry-wide:
At $45b/year (an inflated number, remember!) it's going to take them a long time to recoup the hundreds of billions of dollars they've spent so far. But they don't have a long time: the massive GPUs that power AI's "foundation models" and cost six- or seven-figures each burn out remarkably quickly. The companies that buy these GPUs claim they'll last five years (and depreciate them over that schedule); however, this is accounting fraud, because in reality, these GPUs have a duty-cycle that's more like two to three years:
To recoup their existing and announced investments, AI companies will have to bring in $2 trillion, more than the combined revenue of Amazon, Google, Microsoft, Apple, Nvidia and Meta:
And they have to bring in that $2 trillion before all those GPUs burn out…which is, again, about 2-3 years.
Or sometimes just 54 days.
AI companies' purchases and R&D expenditures aren't guided by the need to make products that will bring in $2 trillion dollars. AI companies spend money in order to put on a show for investors, to demonstrate that they are very serious about AI. Think of all those GPU-stuffed data-centers as akin to a peacock's tailfeathers: an expensive way to attract mates (or, in this case, investors), by emitting costly signals that demonstrate your power:
https://en.wikipedia.org/wiki/Signalling_theory
Of course, it's far cheaper to pretend to be spending a lot of money than it is to actually spend it, and they're doing plenty of that, too. Meta has promised to spend $72b next year on data-centers. However, Meta's annual free cash flow is $52.1b. OpenAI says it will spend $60b/year on data-centers, which is five times its annual revenue of $12.7b (and the company is losing $9b/year). As The American Prospect's Brian McMahon writes, "How can OpenAI plan to spend five times what it brought in?"
I don't know how many of these giant "foundation models" will still be online after the crash, but I would not be surprised if that number is zero.
So the big question is, what comes next? What will the AI bubble leave behind?
Some bubbles leave nothing or next-to-nothing behind. Enron left nothing behind but the cooling corpse of a CEO who popped his clogs before he could be sentenced to life in prison. Worldcom left behind a CEO who survived long enough to die behind bars…and a ton of fiber in the ground that people are still getting use out of (I'm sending these keystrokes to the internet on old Worldcom fiber that AT&T bought and lit up).
Crypto's not going to leave much behind: a few Rust programmers who've really taken security by design to heart, sure, but mostly it'll be shitty Austrian economics and even shittier JPEGs.
So what kind of bubble is AI? That's the $2 trillion question:
Before I get to that, let me be clear here: bubbles are always bad. As much as I like my 2gb symmetrical fiber, the fact that it exists because a crook stole billions of dollars from everyday people who were only hoping to live a dignified retirement of material sufficiency is terrible. Worldcom CEO Bernie Ebbers deserved what he got, and worse.
The AI bubble is on its way to sucking up a trillion dollars and not all of that money is coming from Saudi royals, hedge fund bastards and Elon Musk's credulous creditors. Plenty of it will come out of the savings of working people who've forced to play the suckers at the table thanks to the replacement of guaranteed pensions with "market-based pensions" that only pay out if you guess right about which stocks to buy:
Those people are going to get wrecked. And so are the rest of us. You don't need to be an AI investor to get wiped out by the AI investment bubble, either. With 30+% of the S&P 500 tied up in seven AI companies' stock, the coming crash will definitely escape containment and crash the whole damned economy.
So the bubble is bad. Really bad. But even so, there will be things we can salvage from it: open source models, skilled programmers, cheap GPUs bought out of bankruptcy for pennies on the dollar. It would be better if we created that stuff without burning the world's economy to the ground and emitting a heptillion tons of CO2, but ignoring the productive residue of the AI crash won't bring the economy back, or suck the carbon out of the atmosphere.
The open source models are a big deal. They're already capable of doing really impressive things, like transcription, image generation, and natural language-based data transformation, running on commodity hardware. I run several models on the laptop I'm typing this on – a computer that doesn't even have a GPU.
What's more, there are a lot of ways to improve these models within easy reach. The US AI companies that threw these models over the transom after irrevocably licensing them as free software had very little impetus to improve their efficiency by optimizing them. Remember, they're spending money as a way to "prove" that AI has a future.
Shipping a model that runs badly – that needs more data-centers and energy to run – is a way to convince investors that it's doing something really advanced (after all, look how much compute and energy it's consuming!). It's a scaled-up version of a scam that Elon Musk used to pull on investors when he was shopping his startup Zip2 around: he put the regular PC his demo ran on inside a gigantic hollow case that he would wheel in on a dolly, announcing that his code ran on a "supercomputer." Yes, investors really are that dumb.
Even modest efforts at optimization can yield incredible performance gains. Deepseek, the legendary Chinese open source AI model, consumes a fraction of the resources gobbled up by the likes of OpenAI. Deepseek's launch was so impressive that it knocked $589b off of Nvidia's stock price the day it shipped:
There are a ton of these open source Chinese models, and they all perform like crazy. China does a lot of AI optimization because US embargoes prevent Chinese AI companies from accessing the most powerful GPUs, so Chinese coders tighten up their code and outperform US companies even though they're using far less powerful computers.
After the crash, everyone will be in a similar position to those Chinese AI optimizers: Chinese companies can't buy advanced GPUs because of the embargo; and everyone else won't be able to buy advanced GPUs because the AI crash will have cratered the economy for a generation.
But there is so much room at the bottom. Optimized models do really impressive things on really cheap hardware.
How cheap? Well, here's hardware hacker Pete Warden demoing a chatbot that you talk to and that talks back to you – and it's running on Synaptics System-on-a-Chip (SoC) that costs "low single digit dollars":
This is basically a little special-purpose Alexa, except it doesn't connect to the internet at all (and therefore doesn't leak any of your data). In Warden's demo, the gadget is a button-sized voice assistant that is meant to be integrated into a dishwasher, which can interpret the dishwasher's manual for you. If your dishes come out dirty or if the drain gets clogged, you press the button, describe your issue in pretty vague terms, and it instantly speaks aloud all the troubleshooting steps to deal with it.
This privacy-preserving, cheap-like-borscht component adds a voice-activated, conversational assistant to a device, sipping power like the clock on your microwave, running on a processor that costs less than a pack of AA batteries. It's seriously fucking cool.
There's going to be a lot of this AI, after the AI goes away – just like there was a lot of the web after the dotcom crash, when, overnight, San Francisco had infinity office-space, servers, and techies going begging.
If you'd like an essay-formatted version of this post to read or share, here's a link to it on pluralistic.net, my surveillance-free, ad-free, tracker-free blog:
Maga’s boss class think they are immune to American carnage
If you'd like an essay-formatted version of this post to read or share, here's a link to it on pluralistic.net, my surveillance-free, ad-free, tracker-free blog:
It's not just that Texas DA Gocha Ramirez charged a woman with murder for having an abortion (something he wasn't allowed to do, even under Texas law); it's that Ramirez paid for his mistress's own abortion, after he impregnated her while having an affair with her and her sister:
This is perfect Magaism, as captured by Wilhoit's Law:
Conservatism consists of exactly one proposition, to wit: There must be in-groups whom the law protects but does not bind, alongside out-groups whom the law binds but does not protect.
Maga is a coalition of turkeys voting for Christmas, and ax-sharpening farmers planning to make a meal out of them. The Maga base wants a bunch of stuff that the Maga elites would never tolerate, but that's OK, because the Maga elites are pretty sure they will never have to suffer under the laws they pass for others. Peter Thiel is happy to support a political movement whose dominant factions would like to put him – and every other gay man – in a concentration camp, because he's pretty sure that only applies to the poor gays, not the billionaire gays.
Financiers who back Trump know that they can afford to transport their daughters, wives, mistresses and the housekeepers, babysitters and teenagers they impregnate across state lines (or national borders) to get an abortion should the need arise. Their participation in Maga was a bet that after victory was attained, the base could be made to settle for performative cruelty against people other than them:
The finance sector is the critical faction in Maga, because the financialized ideal is to accumulate wealth and power without exposure to any real-world risks. As Doug Rushkoff writes in Survival of the Richest, the finance move is to "go meta" – don't drive a taxi, buy a medallion and rent it to a taxi driver. Don't buy a medallion, start a rideshare company. Don't start a rideshare company, invest in a rideshare company. Don't invest in a rideshare company, buy options to invest in a rideshare company:
Crypto is as meta as it gets, so no wonder crypto bros are all-in on Trump, and no wonder Trump is all-in on crypto. As Hamilton Nolan writes:
Crypto coins… are pure speculative baubles, endowed with value only to the extent that you can convince another person to pay you more for them than you paid. They are a claim on nothing. They are the grandest embodiment of Greater Fool Theory ever invented by mankind.
Trump's tariffs are blowing up the economy and wiping out the agricultural sector. All those rural, Christmas-voting turkeys are getting it in the neck:
Trump's answer to this is to fire the government statisticians and replace them with work-for-hire fiction hacks who'll publish whatever numbers he tells them to:
You'd think that this would worry the finance sector, but fake numbers are actually good for finance, provided you're on the right side of them. Plenty of people got dynastically rich off of the fake numbers that propped up the pre-2008 housing bubble and the pre-2001 dotcom bubble. Those same people – and their ideological heirs – are now all-in on AI. It's impossible to overstate how structurally important AI is to the US economy. AI bubble companies now account for the value of 35% of the US stock market:
https://www.wheresyoured.at/the-haters-gui/
The instant that bubble pops, the US economy gets a 35% amputation. It's no surprise that, under Trump, the FTC and DoJ have brought the Biden administration's antitrust enforcement against Big Tech to a screeching halt:
Nothing would be worse for the AI bubble than antitrust and securities-law enforcement. Companies that cook their balance sheets and suck up hundreds of billions in investment capital cannot function in a world with an orderly market system overseen by publicly accountable referees charged with keeping everyday people from having their life's savings stolen.
And indeed, Trump's enforcers are running away from their duties, as fast as they can. The latest wheeze is to change the rules so that you can "invest" your retirement savings in cryptocurrency and private equity funds (two tired old swindles whose ropers are scraping the barrel looking for new marks):
Not that AI is much better. AI is hemorrhaging money and bringing in pennies:
https://www.wheresyoured.at/ai-is-a-money-trap/
And things are looking grimmer for AI by the day. It's not just that Openai's latest, "fifth-generation" model was such a spectacular flop that they've been forced to bring back the old version. Far more important is the utter uselessness of AI as a way of realizing cost-savings for the companies that try it:
After all, AI is implicitly a bet on firing workers. The hundreds of billions in investment, the trillions in valuation – these can't be realized by merely making workers' jobs easier or more satisfying. AI isn't a bet on making radiologists better at diagnosing solid-mass lung tumors: it's a bet on firing nearly all the radiologists and using the remainder to be "humans in the loop" for AI, in order to absorb the blame when you die of cancer. There are plenty of radiologists who might welcome AI as a tool they use alongside their traditional workflow – but their bosses aren't about to hand over vast fortunes just to make those workers happier.
This is why AI users often sound like they're using totally different technologies. Workers who get to decide whether and how to incorporate AI into their jobs are doubtless finding lots of utility and delight from the new tool. These workers are "centaurs" – people assisted by machines.
The workers who describe their on-the-job AI as a hellish monstrosity are being ordered to use AI, in workplaces where mass firings have terrified the survivors, who are told they must use the AI to make up for their jobless former colleagues. They are reverse-centaurs: machines assisted by human workers:
There is no way that AI can be worth 35% of the economy if all it does is produce some happy centaurs. The only way that 35% bet pays off is if half the workers get fired and replaced by AI, which is a thing that AI pitchmen are promising, to the letter (a letter that is credulously repeated by the dutiful stenographers of the press):
The problem is that when businesses fire a bunch of workers and replace them with AI, they don't get the promised savings. Instead, they end up with a system that's so broken that all the wage savings are incinerated by the cost of making good on the AI's failures.
But for Maga's finance wing, this is all OK. They're going meta. Don't hire workers, hire AI. Don't hire AI, make AI. Don't make AI, invest in AI. So long as the number keeps going up, finance wins, even if that's only because every structurally important firm in America is being thimblerigged into filling their walls with AI-powered, immortal asbestos that is destined to transform their firms into Superfund sites.
They're betting that when the bubble finally bursts, that they will have become too big to fail, and will thus be in for the bailouts that rescued the finance sector in 2008. They think that so long as they curry favor with Trump, he'll make sure they're all OK, because they are the people the law protects, but does not bind.
This is a pretty good bet. Trump's a gangster capitalist, and fascists love a "dual state" – a system where the law is followed to the letter, except when it suits someone with the protection of the ruling clique to wipe their ass with it:
https://archive.ph/8T8of
And bailouts for finance crooks are a bipartisan consensus. Remember, it was Obama, not Bush, who took his Treasury Secretary Timothy Geithner's advice to allow the bailed-out banks to steal their borrowers homes and trigger the foreclosure crisis, because this would "foam the runways" for the crashing banks:
The Obama wing of the party insists that they're the responsible adults in the room, the ones that will govern wisely and hold their gigadonors to account when they wreck the economy. They tell us Zohran Mamdani is – despite all evidence to the contrary – too unpopular to win an election:
They ratfucked Katie Porter, one of finance's most savage and talented opponents, teaming up with the crypto-bros who are Maga's bagmen. Joke's on them, because it looks like Porter is gonna be California's next governor:
(I donated $100 I can't afford to her campaign; maybe you will donate, too?)
https://secure.actblue.com/donate/kpg_web
Maga's finance wing are convinced that the game is rigged in their favor – heads they win and the law protects them, tails we lose and the law binds us. But if there's one thing we know about gangster capitalism, it's that the capo isn't shy about seizing the fortunes of his various underbosses when the mood suits him. One day he's demanding that you quit your job as CEO, the next day he imposes a 15% tax on your products:
You can bet your ass that if it looks like Trump is gonna lose his grip on power, they'll come sleazing over the Democrats, demanding the defenstration of Mamdani, Porter, and anyone who wants a habitable and just world, rather than a system designed to convert the planet's resources to something that can be sequestered in a luxury bunker or on a private island.
Because for all that they moan about "wokeness," they wouldn't want their kids to have to tolerate a shitty boss; they wouldn't want their kids to carry an unwanted pregnancy to term. They wanna live out their cuckold fantasies in peace:
They don't have any problem with living in a world where there's lip service to social values and Pricewaterhousecooper has a cringe Pride parade float. They'll happily save a couple bucks on the nanny's abortion by going down to the corner Planned Parenthood rather than flying her to Toronto on the private jet. All that performative cruelty was just a shuck to get some of the dumber surviving turkeys to pull the lever for Christmas. So long as they can live in a world where the law protects them, but does not bind them, they're happy as pigs in shit.
Hey, German-speakers! Through a very weird set of circumstances, I ended up owning the rights to the German audiobook of my bestselling 2022 cryptocurrency heist technothriller Red Team Blues and now I'm selling DRM-free audio and ebooks, along with the paperback (all in German and English) on a Kickstarter that runs until August 11.
Not only is agentic AI bullshit, but it's a specific kind of bullshit that AI hucksters have busted out in the past, and will bust out in the future, so it's worth spending a minute to unpack this bullshit and catalog its traits so that we don't fall for it. As GW Bush says, "Fool me once, shame on you; fool me twice, we don't get fooled again."
Automation can be transformative, relieving us of danger and drudgery by getting a machine to pick up some of the heavy work. Ideally automation seamlessly swaps a human for a machine at some stage in a process (ideally, the boring, dangerous and/or difficult phase). Like, whipping egg-whites for a meringue is hard on your wrist. But swap your whisk for a hand blender, and suddenly that tiresom process becomes fast and easy. If the blender is cordless, you can use it anywhere in your kitchen, including wherever you would have stood over a bowl with a whisk.
A mixer, by contrast, requires more labor on your part: you have to decant the contents of your mixing bowl into the mixer, run its motor, and then scrape the whipped whites back into your bowl for the next phase. It's worse automation.
But the worst automation would be a mixer that requires a special electrical outlet, a different fridge, and a special egg-carton. You would have to redesign your whole kitchen to use that thing. Sure, it might produce perfect meringues, and sure, if you had a meringue factory it might be a great solution. But for everyday use, it's a solution that creates more problems than it solves.
AI pitchmen promise that seamless swapping of a human tethered to some choresome drudgery for software. That's the whole point of self-driving cars: each of us can swap a standard car for one with an autopilot and use the same roads, with the same road-users, to get to all the same places. We don't have to tear up all the roads and lay tracks, or fill the roadside environment with sensors and beacons to help the "self-driving" cars navigate the system. A self-driving car can share the road with human-piloted vehicles, even when those other vehicles are driven by humans who don't see why they should allow a robot to merge into their lane or have the right of way, even if the human is turning left into oncoming robo-traffic.
Self-driving cars are not very good at this stuff, as it turns out. When that became apparent, self-driving car hucksters announced that it was only reasonable for their products to require something of the rest of us. As Andrew Ng put it:
“I think many AV teams could handle a pogo stick user in pedestrian crosswalk,” Ng told me. “Having said that, bouncing on a pogo stick in the middle of a highway would be really dangerous.”
“Rather than building AI to solve the pogo stick problem, we should partner with the government to ask people to be lawful and considerate,” he said. “Safety isn’t just about the quality of the AI technology.”
This is an incredible act of shameless bait-and-switchery. In just a few short sentences, Ng's cars go from being the kind of automation that is purely the concern of the person who uses it – the owner of a self-driving car – to the kind of automation that everyone in the world has to adjust to, lest we become part of the "pogo stick problem."
Making a car that can navigate a well-behaved, non-adversarial world is relatively straightforward. But demanding that the entire world behave itself? Well, that's the hard problem of 100,000 years of civilization and ethics. A product that only works in an ideal world isn't a viable product.
Self-driving car boosters didn't invent this wheeze, either. The entire concept of "pedestrian" (and later, "jaywalker") was invented by the auto industry to shift blame for the death and destruction the wealthy owners of their products inflicted on everyday people to the victims:
The latest peddlers of pogo-stick demands are the agentic AI people. They have raised (hundreds of) billions of dollars by promising that they will make AIs that can autopilot your browser to accomplish tedious, time-consuming tasks, visiting the same websites you would visit, locating and processing the information needed to perform the task you've set for it. This will supposedly make all kinds of human workers obsolete (which is where the hundreds of billions of dollars come in – the whole AI investor pitch is "We are developing technology that will let bosses fire their workers").
But agentic AI sucks. Asking a chatbot to take a screenshot of a website, then make guesses about which parts of it are links and what those links do, choose one link to fire a click at, and then start again is a recipe for incredible dysfunction. That's even before we get into "hallucinations" (this is AI jargon for "errors").
A more mature agentic AI apologetics admits that while no one knows how to make an AI that can navigate the whole internet, we can make specialist agents that can perform one kind of task, then hand off the output from that task to the next agent, and the next. This also sucks: you're created a whole menagerie of AIs, each of which is prone to its own failure modes, and then combining them, multiplying all those error potentials together, sending erroneous findings careening through a cascade of downstream AIs. This is broken-telephone-as-a-service. Give it your credit card, ask it to order a bag of jucing oranges, and six months later someone's gonna back a 16 wheeler up to your front door with $40,000 worth of frozen OJ and a receipt for a futures contract you're on the hook for.
The latest agentic AI pitch "solves" this problem by asserting that the whole internet will simply have to accommodate itself to AI agents. Every website will have to adopt robust, accurate semantics that describe its navigation and offerings, standardized across every domain of human activity. This would be great. The semantic web people have been trying to make it happen since 1999, with no success to speak of, for reasons I identified more than 20 years ago:
The reason websites don't make their results easy to scrape and compare is that they want to cheat you. They want you to buy something more expensive and/or inferior than the best match for your desire. There is no way for an AI agent to know when a website is lying to it, and the websites that lie the most are incentivized to have the best, highest-grade automation hooks for an AI agent to connect to (just as spammers have the best, most pristine anti-spam indicia, from DKIM to SPF to DMARC records).
And these cheaters aren't fringe players – they're the biggest companies out there. Amazon knows that Prime members don't shop around, so it presents them with higher prices than non-Prime users. Airlines use AI and surveillance data to estimate your desperation and price their tickets accordingly:
The hard part of comparison-shopping for an airline isn't sorting a database of all the prices offered to all customers under all circumstances: it's compiling such a database. We don't need complex AI-based techniques to perform a simple sort – we need AI to solve the problem of knowing what prices every airline is charging at this instant to every flier for every itinerary.
When agentic AI grifters insist that the entire internet has to adopt and faithfully use standard APIs so their bots can accurately analyze the internet's contents, they are re-inventing the pogo-stick problem. Yes, if you could get the entire world to arrange its affairs to your benefit, you could surely do some incredible things, and if my grandmother had wheels, she'd be a rollerskate.
Even if you could get everyone to adopt a standard set of APIs and use them well, this is a titanic engineering challenge, at least as big as anything the agentic AI people are promising to do.
There's an unassailable response to the assertion that you could do amazing things as soon as everyone else upends their life to make things more convenient for you, the sacred principle of "wish in one hand and shit in the other and see which will be full first":
Support me this summer in the Clarion Write-A-Thon and help raise money for the Clarion Science Fiction and Fantasy Writers' Workshop! This summer, I'm writing The Reverse-Centaur's Guide to AI, a short book for Farrar, Straus and Giroux that explains how to be an effective AI critic.
If you'd like an essay-formatted version of this post to read or share, here's a link to it on pluralistic.net, my surveillance-free, ad-free, tracker-free blog:
I'm coming to COLORADO! Catch me in DENVER on Jan 22 at The Tattered Cover<, and in COLORADO SPRINGS from Jan 23–25 where I'm the Guest of Honor at COSine. Then I'll be in OTTAWA on Jan 28 at Perfect Books and in TORONTO with Tim Wu on Jan 30.
Like all the best Americans, I'm Canadian, and while I have lived abroad for most of this century, I still hew faithfully to our folkways, which is why I'd like to start this essay by apologizing.
I'm sorry.
I'm sorry! I'm a technology writer, which means I'm supposed to be encouraging you to throw hundreds of billions of dollars at the money-losingest technology in human history, AI. No one has ever lost as much money as the AI companies.
There is no way to operate one of Nvidia's big AI-optimized GPUs without losing money. The owners of these GPUs who have lost the least money are the ones who rushed into buying GPUs without ensuring they'd have electricity to power them, and have been forced to leave their GPUs to age in warehouses. The minute they plug in those GPUs, they'll start losing money, and the more they use them, the more money they'll lose.
I'm sorry. As a technology writer, I'm supposed to be telling you that this bet will some day pay off, because one day we will have shoveled so many words into the word-guessing program that it wakes up and learns how to actually do the jobs it is failing spectacularly at today. This is a proposition akin to the idea that if we keep breeding horses to run faster and faster, one of them will give birth to a locomotive. Humans possess intelligence, and machines do not. The difference between a human and a word-guessing program isn't how many words the human knows.
I'm sorry. I know that when we talk about "digital sovereignty," we're obliged to talk about how we can build more data-centres that we can fill up with money-losing chips from American silicon monopolists in the hopes of destroying as many jobs as possible while blowing through our clean energy goals and enshittifying as much of our potable water as possible.
I don't have any advice for how to do that. I'm sorry!
As Canada contemplates our response to the collapse of the American empire and its alliances with the world, the cornerstone of our current strategy is sacrificing our dollars, water and energy in order to become more dependent on America, in a weird and improbable bet that we will figure out how to make millions of Canadians unemployed. I'm sorry, that just doesn't sound like a great idea to me.
If I can beg your indulgence, I'd like to propose an alternative.
Back in 2012, Canada passed Bill C-11, the Copyright Modernization Act. It's a law that bans Canadian companies from modifying America's digital tech exports. We passed it because the US threatened us with tariffs:
Thanks to Bill C-11, a Canadian company can't sell jailbreaking kits for phones and consoles, which would let Canadian sellers offer goods and services to Canadian buyers outside of US app stores, sidestepping the 30% app tax that Apple, Google, Microsoft, Sony and others impose on our digital economy.
Thanks to Bill C-11, a Canadian company can't sell mechanics a universal diagnostic tool that turns every "check engine" light into a useful error message. Instead, Canadian mechanics have to send $10,000/year/manufacturer to America for a proprietary car diagnosis kit.
Thanks to Bill C-11, a Canadian company can't offer ink cartridge manufacturers software that will ensure their cartridges work in the printers Canadians buy from the American inkjet cartel. As a result, Canadians have to spend $10,000/gallon on ink, making it the most expensive fluid a Canadian civilian can purchase without a government permit.
Thanks to Bill C-11, a Canadian company can't sell our farmers software that lets them start using their tractors as soon as they've fixed them. Instead, after a Canadian farmer fixes their tractor, they have to wait for a service call from a rep for a US ag-tech monopolist who'll type an unlock code into the tractor's keyboard and charge the farmer a couple hundred bucks for this "service."
Thanks to Bill C-11, a Canadian company can't revive one of the most successful technologies in modern history: the home video recorder. Remember those? First we had VCRs, then we had digital successors like the Tivo. Canadian law says you're allowed to record the video that comes into your home, whether by broadcast, cable, satellite or streaming. But Bill C-11 bans a Canadian company from selling you a gadget that lets you save the video you get in an app or from a set-top box.
It's crazy: we have actually uninvented the VCR! You know how everyone is pissed off about their favourite shows being yanked from the streaming services? Repeal C-11 and you could just save those shows forever. Repeal C-11 and you'd kill the grinchy little racket that services like Prime pull, where Christmas cartoons are in the free tier from March to November, and cost $3.99 to watch between November and March. Just tape 'em in August and save 'em for later!
It doesn't stop there. Remember when Facebook banned all links to the news in Canada? Repeal C-11 and a Canadian company could sell you an alternative Facebook app that puts the news back into your feed! Repeal C-11 and Canadians could get an alternative app that replaces all the streaming services, letting you search and stream every service you have an account for in one place, mixing in Canadian content from the NFB, public broadcasters, and commercial services.
Virtually every Canadian ministry, corporation and household is locked into a US Big Tech silo. Any of these could be shut down at a single word from Trump to any of the tech giants who've lined up to do his bidding. Repeal C-11 and we can extract all our data from these walled gardens/prisons and get it onto auditable, trustworthy, transparent open source software, hosted in data-centres located safely on Canadian soil.
If there's one thing Canadians are good it, it's going to other countries and extracting their wealth. We're world champions at it.
America's tech monopolies have sequestered trillions of dollars worth of monopoly rents on their balance sheets. This is dead capital, being pissed up the wall on nonsense like stock buybacks and data-centres and grotesque executive bonuses.
As Jeff Bezos said to the publishers: "Your margin is my opportunity."
America's tech trillions represent a rich and readily accessible seam that we can extract – safely, from our own country! – and turn into our billions, and an exportable line of products that the whole world would beat a path to our door to buy.
Look, I'm sorry. I don't have any ideas for how Canada can get to a better future by lighting billions on fire in a bet on a failing technology whose dubious profitability depends on ruining our job market, our power grid and our water supply, which will tie the American political situation to our ankles.
All I've got is an idea for how we can make insanely profitable products that people really want to buy, that will insulate us from cyberattacks by US tech giants who are in thrall to Trump, and that Americans will pay us to use in order to free themselves from the tech giants who abuse them, too.
I'm really sorry. I know it's out of step with the times, but all I have is ideas that make money, make us safer, make us richer, and make our technology better.
On the other hand, those chatbots sure are cute. It's funny when they "hallucinate."
If you'd like an essay-formatted version of this post to read or share, here's a link to it on pluralistic.net, my surveillance-free, ad-free, tracker-free blog:
My latest column for Locus Magazine is "What Kind of Bubble is AI?" All economic bubbles are hugely destructive, but some of them leave behind wreckage that can be salvaged for useful purposes, while others leave nothing behind but ashes:
Think about some 21st century bubbles. The dotcom bubble was a terrible tragedy, one that drained the coffers of pension funds and other institutional investors and wiped out retail investors who were gulled by Superbowl Ads. But there was a lot left behind after the dotcoms were wiped out: cheap servers, office furniture and space, but far more importantly, a generation of young people who'd been trained as web makers, leaving nontechnical degree programs to learn HTML, perl and python. This created a whole cohort of technologists from non-technical backgrounds, a first in technological history. Many of these people became the vanguard of a more inclusive and humane tech development movement, and they were able to make interesting and useful services and products in an environment where raw materials – compute, bandwidth, space and talent – were available at firesale prices.
Contrast this with the crypto bubble. It, too, destroyed the fortunes of institutional and individual investors through fraud and Superbowl Ads. It, too, lured in nontechnical people to learn esoteric disciplines at investor expense. But apart from a smattering of Rust programmers, the main residue of crypto is bad digital art and worse Austrian economics.
Or think of Worldcom vs Enron. Both bubbles were built on pure fraud, but Enron's fraud left nothing behind but a string of suspicious deaths. By contrast, Worldcom's fraud was a Big Store con that required laying a ton of fiber that is still in the ground to this day, and is being bought and used at pennies on the dollar.
AI is definitely a bubble. As I write in the column, if you fly into SFO and rent a car and drive north to San Francisco or south to Silicon Valley, every single billboard is advertising an "AI" startup, many of which are not even using anything that can be remotely characterized as AI. That's amazing, considering what a meaningless buzzword AI already is.
So which kind of bubble is AI? When it pops, will something useful be left behind, or will it go away altogether? To be sure, there's a legion of technologists who are learning Tensorflow and Pytorch. These nominally open source tools are bound, respectively, to Google and Facebook's AI environments:
But if those environments go away, those programming skills become a lot less useful. Live, large-scale Big Tech AI projects are shockingly expensive to run. Some of their costs are fixed – collecting, labeling and processing training data – but the running costs for each query are prodigious. There's a massive primary energy bill for the servers, a nearly as large energy bill for the chillers, and a titanic wage bill for the specialized technical staff involved.
Once investor subsidies dry up, will the real-world, non-hyperbolic applications for AI be enough to cover these running costs? AI applications can be plotted on a 2X2 grid whose axes are "value" (how much customers will pay for them) and "risk tolerance" (how perfect the product needs to be).
Charging teenaged D&D players $10 month for an image generator that creates epic illustrations of their characters fighting monsters is low value and very risk tolerant (teenagers aren't overly worried about six-fingered swordspeople with three pupils in each eye). Charging scammy spamfarms $500/month for a text generator that spits out dull, search-algorithm-pleasing narratives to appear over recipes is likewise low-value and highly risk tolerant (your customer doesn't care if the text is nonsense). Charging visually impaired people $100 month for an app that plays a text-to-speech description of anything they point their cameras at is low-value and moderately risk tolerant ("that's your blue shirt" when it's green is not a big deal, while "the street is safe to cross" when it's not is a much bigger one).
Morganstanley doesn't talk about the trillions the AI industry will be worth some day because of these applications. These are just spinoffs from the main event, a collection of extremely high-value applications. Think of self-driving cars or radiology bots that analyze chest x-rays and characterize masses as cancerous or noncancerous.
These are high value – but only if they are also risk-tolerant. The pitch for self-driving cars is "fire most drivers and replace them with 'humans in the loop' who intervene at critical junctures." That's the risk-tolerant version of self-driving cars, and it's a failure. More than $100b has been incinerated chasing self-driving cars, and cars are nowhere near driving themselves:
Quite the reverse, in fact. Cruise was just forced to quit the field after one of their cars maimed a woman – a pedestrian who had not opted into being part of a high-risk AI experiment – and dragged her body 20 feet through the streets of San Francisco. Afterwards, it emerged that Cruise had replaced the single low-waged driver who would normally be paid to operate a taxi with 1.5 high-waged skilled technicians who remotely oversaw each of its vehicles:
The self-driving pitch isn't that your car will correct your own human errors (like an alarm that sounds when you activate your turn signal while someone is in your blind-spot). Self-driving isn't about using automation to augment human skill – it's about replacing humans. There's no business case for spending hundreds of billions on better safety systems for cars (there's a human case for it, though!). The only way the price-tag justifies itself is if paid drivers can be fired and replaced with software that costs less than their wages.
What about radiologists? Radiologists certainly make mistakes from time to time, and if there's a computer vision system that makes different mistakes than the sort that humans make, they could be a cheap way of generating second opinions that trigger re-examination by a human radiologist. But no AI investor thinks their return will come from selling hospitals that reduce the number of X-rays each radiologist processes every day, as a second-opinion-generating system would. Rather, the value of AI radiologists comes from firing most of your human radiologists and replacing them with software whose judgments are cursorily double-checked by a human whose "automation blindness" will turn them into an OK-button-mashing automaton:
The profit-generating pitch for high-value AI applications lies in creating "reverse centaurs": humans who serve as appendages for automation that operates at a speed and scale that is unrelated to the capacity or needs of the worker:
But unless these high-value applications are intrinsically risk-tolerant, they are poor candidates for automation. Cruise was able to nonconsensually enlist the population of San Francisco in an experimental murderbot development program thanks to the vast sums of money sloshing around the industry. Some of this money funds the inevitabilist narrative that self-driving cars are coming, it's only a matter of when, not if, and so SF had better get in the autonomous vehicle or get run over by the forces of history.
Once the bubble pops (all bubbles pop), AI applications will have to rise or fall on their actual merits, not their promise. The odds are stacked against the long-term survival of high-value, risk-intolerant AI applications.
The problem for AI is that while there are a lot of risk-tolerant applications, they're almost all low-value; while nearly all the high-value applications are risk-intolerant. Once AI has to be profitable – once investors withdraw their subsidies from money-losing ventures – the risk-tolerant applications need to be sufficient to run those tremendously expensive servers in those brutally expensive data-centers tended by exceptionally expensive technical workers.
If they aren't, then the business case for running those servers goes away, and so do the servers – and so do all those risk-tolerant, low-value applications. It doesn't matter if helping blind people make sense of their surroundings is socially beneficial. It doesn't matter if teenaged gamers love their epic character art. It doesn't even matter how horny scammers are for generating AI nonsense SEO websites:
These applications are all riding on the coattails of the big AI models that are being built and operated at a loss in order to be profitable. If they remain unprofitable long enough, the private sector will no longer pay to operate them.
Now, there are smaller models, models that stand alone and run on commodity hardware. These would persist even after the AI bubble bursts, because most of their costs are setup costs that have already been borne by the well-funded companies who created them. These models are limited, of course, though the communities that have formed around them have pushed those limits in surprising ways, far beyond their original manufacturers' beliefs about their capacity. These communities will continue to push those limits for as long as they find the models useful.
These standalone, "toy" models are derived from the big models, though. When the AI bubble bursts and the private sector no longer subsidizes mass-scale model creation, it will cease to spin out more sophisticated models that run on commodity hardware (it's possible that Federated learning and other techniques for spreading out the work of making large-scale models will fill the gap).
So what kind of bubble is the AI bubble? What will we salvage from its wreckage? Perhaps the communities who've invested in becoming experts in Pytorch and Tensorflow will wrestle them away from their corporate masters and make them generally useful. Certainly, a lot of people will have gained skills in applying statistical techniques.
But there will also be a lot of unsalvageable wreckage. As big AI models get integrated into the processes of the productive economy, AI becomes a source of systemic risk. The only thing worse than having an automated process that is rendered dangerous or erratic based on AI integration is to have that process fail entirely because the AI suddenly disappeared, a collapse that is too precipitous for former AI customers to engineer a soft landing for their systems.
This is a blind spot in our policymakers debates about AI. The smart policymakers are asking questions about fairness, algorithmic bias, and fraud. The foolish policymakers are ensnared in fantasies about "AI safety," AKA "Will the chatbot become a superintelligence that turns the whole human race into paperclips?"
But no one is asking, "What will we do if" – when – "the AI bubble pops and most of this stuff disappears overnight?"
If you'd like an essay-formatted version of this post to read or share, here's a link to it on pluralistic.net, my surveillance-free, ad-free, tracker-free blog:
I'm on a 20+ city book tour for my new novel PICKS AND SHOVELS. Catch me in AUSTIN on Mar 10. I'm also appearing at SXSW and at many events around town, for Creative Commons, Fediverse House, and EFF-Austin. More tour dates here.
Big Tech's astonishing scale is matched only by its farcical valuations – price-to-earnings ratios that consistently dwarf the capitalization of traditional hard-goods businesses. For example, Amazon's profit-to-earnings ratio is 37.65; Target's is only 13.34. That means that investors value every dollar Amazon brings in at three times the value they place on a dollar spent at Target.
The fact that Big Tech stocks trade at such a premium isn't merely of interest to tech investors, or even to the personal wealth managers who handle the assets of tech executives whose personal portfolios are full of their employers' stock options.
The high valuations of tech stocks don't just reflect an advantage over bricks and mortar firms – they are the advantage. If you're Target and you're hoping to hire someone who's just interviewed at Amazon, you have to beat Amazon's total compensation offer. But when Amazon makes that offer, they can pay some – maybe even most – of the offer in stock, rather than in cash.
This is a huge advantage! After all, to get dollars, both Amazon and Target have to convince you to spend money in their stores (or, in Amazon's case, with its cloud, or as a Prime sub, etc etc). Both Amazon and Target get their dollars from entities outside of the firm's four walls, and the dollars only come in when they convince someone else to do business with them.
But stock comes from inside the firm. Amazon makes new Amazon shares by typing zeroes into a spreadsheet. They don't have to convince you to buy anything in order to issue that new stock. That is their call, and their call alone.
Amazon can buy lots of things with stock – not just the labor of in-demand technical workers who command six-figure salaries. They can even buy whole companies using stock. So if Amazon and Target are bidding against one another for an anticompetitive acquisition of a key supplier or competitor, Amazon can beat Target's bid without having to spend the dollars its shareholders would like them to divert to dividends, stock buybacks, etc.
In other words, a company with a fantastic profit/earning ratio has its own money-printer that produces currency that can be used to buy labor and even acquire companies.
But why do investors value tech stocks so highly? In part, it's just circular reasoning: a company with a high stock price can beat its competitors because it has a high stock price, so I should buy its stock, which will drive up its stock price even further.
But there's more to this than self-fulfilling prophecy. The high price of tech stocks reflects the market's belief that these companies will continue to grow. If you think a company will be ten times bigger in two years, and it's only priced at three times as much as mature rivals that have stopped growing altogether, then that 300% stock premium is a bargain, because the company will have 1,000% growth in just a couple years. Tech companies have proven themselves, time and again, to be capable of posting incredible growth – think of how quickly Google went from a niche competitor to established search engines to the dominant player, with a 90% market share.
That kind of growth is enough to make anyone giddy, but it eventually runs up against the law of large numbers: doubling a small number is easy, doubling a large number is much, much harder. A search engine that's used by 90% of the world can't double its users – there just aren't enough people to sign up. They'd need to breed several billion new humans, raise them to maturity, and then convince them to be Google users.
And here's the thing: the flipside of the huge profits that can be reaped by investors who buy stocks at a premium in anticipation of growth is the certainty that you will be wiped out if you're still holding the stock when the growth halts. When Amazon stops growing, its PE ratio should fall to something like Target's, which means that its stock should decline by two thirds on that day.
Which is why Big Tech investors tend to be twitchy, hair-trigger types, easily stampeded into mass selloffs. That's what happened in 2022, when Facebook admitted to investors that it had grown more slowly than it had projected, and investors staged the largest stock selloff in history (to that point – hi, Nvidia!), wiping a quarter-trillion dollars off Meta's valuation in a day:
As Stein's Law has it: "anything that can't go on forever eventually stops." Growth stocks have to stop growing, eventually, and when they do, you'd better beat everyone else to the fire exit, or you're going to get crushed in the stampede.
Which is why tech companies are so obsessed with both actual growth, and stories about growth. Facebook spent tens of billions on bribes to telcos around the world, demanding that they charge extra to access non-Facebook websites and apps, in a bid to sign up "the next billion users":
That wasn't just about some ideological commitment to growth – it was about the real, material advantages that a growing company has, namely, that it can substitute the stock it creates for free by typing zeroes into a spreadsheet for money that it can only get by convincing you to give your money to it.
"Facebook Zero" (as this bribery program was called) was about actual growth: finding people who weren't Facebook users and turning them into Facebook users, preferably forever (thanks to Facebook's suite of lock-in tactics that make it a digital roach motel that users check into but don't check out of):
But plenty of the things that Big Tech gets up to are about the narrative of growth. That's why Big Tech has pumped every tech bubble of this stupid decade: metaverse, cryptocurrency, AI. These technologies have each been at the forefront of Big Tech marketing and investor communications, but not solely because they represented a market opportunity. Rather, they represented a more-or-less plausible explanation for how these companies that were on the wrong side of the law of large numbers could continue to double in size, without breeding billions of new customers to sign up for their services.
The tell – as always – comes in the way that these companies refute their critics. When critics point out that Facebook spent $1.2 billion on a metaverse product that only has 32 users:
Or that hardly anyone uses AI, and what uses it does have are often low-value:
https://www.wheresyoured.at/oai-business/
The "narrative entrepreneurs" behind the claims of infinite growth from these technologies all have the same response: "That's what they said about the web, and yet it grew really fast! People who lacked the vision to understand the web's potential missed out. Buy [crypto|metaverse|AI] or have fun being poor!"
It's true – there were a lot of people who were blithely dismissive of the web, and they were wrong. But the fact that the web's skeptics were wrong doesn't mean that skepticism itself is foolish. People were also skeptical of Qibi, Beanie Babies, and the Segway – all of which were predicted to continue to increase in value forever and become permanently installed as significant facts in the economy. The fact that lots of people think something is stupid is not a reliable indicator that it is actually great.
So it's not just that capitalism adopts "the ideology of a tumor" in insisting that infinite growth is possible. The value in corporate claims to eternal growth is not aesthetic, it is material. If the market believes a company will grow, then that company gets to print its own money, which lets it outcompete mature rivals, which lets it grow some more.
But! When the company runs out of growth potential, the process runs in reverse. Not only do executives – whose portfolios are stuffed full of their own company's shares – stand to lose most of their net worth overnight, but once a company's stock starts to decline, it can expect to see an exodus of the key personnel who are compensated in now-worthless stock. That means that once a company hits a bad bump in the road that sets it off course, it needs to worry about losing all the skilled employees who can get it back on the road.
So growth is important, not for its own sake, but for how it affects the cost basis of companies, and thus determines their competitive outlook. But not all growth is created equal.
Remember when Facebook pissed away billions in a bid to capture "the next billion users"? Those users – people from poor countries in the global south – were not as valuable to Facebook as its US customers. The news that sparked a $250 billion, one-day selloff of Facebook shares wasn't merely about anemic growth – it was specifically about anemic growth in the USA.
American customers are worth more than other users to Big Tech – that's true even of users from other populous countries, and of users from other wealthy countries. Norway is rich as hell, but each Norwegian Facebook user is worth pennies on the kroner compared to American users. And there are brazilians of people in South America, but they're worth even less per capita than Norwegians are. Even the whole EU, with its 500m+ relatively wealthy consumers, is only worth a fraction of the US market.
Why is the American market so prized by Big Tech? Because it the only country in the world at the center of a Venn diagram with three overlapping circles. America is the only country in the world that is:
a) populous;
b) wealthy; and
c) totally lacking in legal privacy protections.
The US Congress last updated American consumer privacy law in 1988, when the Video Privacy Protection Act was passed to protect Americans from the high-tech threat of…video store clerks leaking your rental history to the newspapers. Despite the bewildering, obvious, serious privacy risks that have emerged since Die Hard was in theaters, Congress has done nothing to extend Americans' consumer privacy rights.
There are other rich countries where privacy law sucks, but they are small countries with few people. There are extremely populous poor countries with shitty privacy laws, but they're poor. Tech has to steal the private data of dozens of those people to make as much money as they can get from selling the data of just one American. And there are other rich, populous countries – like Germany, say – but those countries actually defend the privacy of the people who live there, and so the revenue tech gets from each of those users is even lower than the RPU for the undefended poor people of the global south.
America is exceptional in that it represents the one place where there are lots of wealthy people who are totally defenseless. We're an all-you-can-eat buffet for the privacy-annihilating voyeurs of Silicon Valley.
These are the two dirty secrets of Big Tech's economics. These companies are reliant on the fragile narrative of infinite growth, and that narrative isn't merely about global growth, but it is particularly and especially about growth in the USA.
Tech's power comes from an implausible story of discovering an endless stream of Americans to sign up and screw over. That story is extremely load-bearing – so much so that by the instant at which the first crack appears, collapse is only moments away. And boy, are there cracks:
https://www.wheresyoured.at/power-cut/
If you'd like an essay-formatted version of this post to read or share, here's a link to it on pluralistic.net, my surveillance-free, ad-free, tracker-free blog:
If you'd like an essay-formatted version of this post to read or share, here's a link to it on pluralistic.net, my surveill ance-free, ad-free, tracker-free blog:
Postmortems and blame for the 2024 elections are thick on the ground, but amidst all those theories and pointed fingers, one explanation looms large and credible: the American housing emergency. If the system can't put a roof over your head, that system needs to go.
American housing has been in crisis for decades, of course, but it keeps getting worse…and worse…and worse. Americans pay more for worse housing than at any time in their history. Homelessness is at a peak that is soul-crushing to witness and maddening to experience. We turned housing – a human necessity second only to air, food and water – into an asset governed almost entirely by market forces, and so created a crisis that has consumed the nation.
The Trump administration has no plan to deal with housing. Or rather, they do have plans, but strictly of the "bad ideas only" variety. Trump wants to deport 11m undocumented immigrants, and their families, including citizens and Green Card holders (otherwise, that would be "family separation" and that's cruel). Even if you are the kind of monster who can set aside the ghoulishness of solving your housing problems by throwing someone in a concentration camp at gunpoint and then deporting them to a country where they legitimately fear for their lives, this still doesn't solve the housing emergency, and will leave America several million homes short.
Their other solution? Deregulation and tax cuts. We've seen this movie before, and it's an R-rated horror flick. Financial deregulation created the speculative mortgage markets that led to the 2008 housing crisis, which created a seemingly permanent incapacity to build new homes in America, as skilled tradespeople retired or changed careers and housebuilding firms left the market. Handing giant tax cuts to the monopolists who gobbled up the remains of these bankrupt small companies minted a dozen new housing billionaires who preside over companies that make more money than ever by building fewer homes:
This isn't working. Homelessness is ballooning. The only answer Trump and his regime have for our homeless neighbors is to just make it a crime to be homeless, sweeping up homeless encampments and busting homeless people for "loitering" (that is, existing in space). There is no universe in which this reduces homelessness. People who lose their homes aren't going to dig holes, crawl inside, and pull the dirt down on top of themselves. If anything, sweeps and arrests will make homelessness worse, by destroying the possessions, medication and stability that homeless people need if they are to become housed.
Today, The American Prospect published an excellent package on the housing emergency, looking at its causes and the road-tested solutions that can work even when the federal government is doing everything it can to make the problem worse:
The Harris campaign ran on Biden's economic record, insisting that he had tamed inflation. It's true that the Biden admin took action against monopolists and greedflation, including criminal price-fixing companies like Realpage, which helps landlords coordinate illegal conspiracies to rig rents. Realpage sets the rents for the majority of homes in major metros, like Phoenix:
Of course, reducing inflation isn't the same as bringing prices down – it just means prices are going up more slowly. And sure, inflation is way down in many categories, but not in housing. In housing, inflation is accelerating:
The housing emergency makes everything else worse. Blue states are in danger of losing Congressional seats because people are leaving big cities: not because they want to, but because they literally can't afford to keep a roof over their heads. LGBTQ people fleeing fascist red state legislatures and their policies on trans and gay rights can't afford to move to the states where they will be allowed to simply live:
So what are the roots of this problem, and what can we do about it? The housing emergency doesn't have a unitary cause, but among the most important factors is fuckery that led to the Great Financial Crisis and the fuckery that followed on from it, as Ryan Cooper writes:
The Glass-Steagall Act was a 1933 banking regulation created to prevent Great Depression-style market crashes. It was killed in 1999 by Bill Clinton, who declared, "the Glass–Steagall law is no longer appropriate." Nine years later, the global economy melted down in a Great Depression-style market crash fueled by reckless speculation of the sort that Glass-Steagall had prohibited.
The crash of 2008 took down all kinds of industries, but none were so hard-hit as home-building (after all, mortgages were the raw material of the financial bubble that popped in 2008). After 2008, construction of new housing fell by 90% for the next two years. This protracted nuclear winter in the housing market killed many associated industries. Skilled tradespeople retrained, or "left the job market" (a euphemism for becoming disabled, homeless, or destroyed). Waves of bankruptcies swept through the construction industry. The construction workforce didn't recover to pre-crisis levels for 16 years (and of course, by then, there was a huge backlog of unbuilt homes, and a larger population seeking housing).
Meanwhile, the collapse of every part of the housing supply chain – from raw materials to producers – set the stage for monopoly rollups, with the biggest firms gobbling up all these distressed smaller firms. Thanks to this massive consolidation, homebuilders were able to build fewer houses and extract higher profits by gouging on price. They doubled down on this monopoly price-gouging during the pandemic supply shocks, raising prices well above the pandemic shortage costs.
The housing market is monopolized in ways that will be familiar to anyone angry about consolidation in other markets – from eyeglasses to pharma to tech. One builder, HR Horton, is the largest player in 3 of the country's largest markets, and it has tripled its profits since 2005 while building half as many houses. Modern homebuilders don't build: they use their scale to get land at knock-down rates, slow-walk the planning process, and then farm out the work to actual construction firms at rates that barely keep the lights on:
Monopolists can increase profits by constraining supply. 60% of US markets are "highly concentrated" and the companies that dominate these markets are starving homebuilding in them to the tune of $106b/year:
There are some obvious fixes to this, but they are either unlikely under Trump (antitrust action to break up builders based on their share in each market) or impossible to imagine (closing tax loopholes that benefit large building firms). Likewise, we could create a "homes guarantee" that would act as an "automatic stabilizer." That would mean that any time the economy slips into recession, this would trigger automatic funding to pay firms to build public housing, thus stimulating the economy and alleviating the housing supply crisis:
The Homes Guarantee is further explained in a separate article in the package by Sulma Arias from People's Action, who describes how grassroots activists fighting redlining planted the seeds of a legal guarantee of a home:
Arias describes the path to a right to a home as running through the mass provision of public housing – and what makes that so exciting is that public housing can be funded, administered and built by local or state governments, meaning this is a thing that can happen even in the face of a hostile or indifferent federal regime.
In Paul E Williams's story on FIMBY (finance in my back yard), the executive director of Center for Public Enterprise offers an inspirational story of how local governments can provide thousands of homes:
Williams recounts the events of 2021 in Montgomery County, Maryland, where a county agency stepped in to loan money to a property developer who had land, zoning approval and work crews to build a major new housing block, but couldn't find finance. Montgomery County's Housing Opportunities Commission made a short-term loan at market rates to the developer.
By 2023, the building was up and the loan had been repaid. All 268 units are occupied and a third are rented at rates tailored to low-income tenants. The HOC is the permanent owner of those homes. It worked so well that Montgomery's HOC is on track to build 3,000 more public homes this way:
Other – in red states! – have followed suit, with lookalike funds and projects in Atlanta and Chattanooga, with "dozens" more plans underway at state and local levels. The Massachusetts Momentum Fund is set to fund 40,000 homes.
The Center for Public Enterprise has a whole report on these "Government Sponsored Enterprises" and the role they can play in creating a supply of homes priced at a rate that working people can afford:
Of course, for a GSE to loan money to build a home, that home has to be possible. YIMBYs are right to point to restrictive zoning as a major impediment to building new homes, and Robert Cruickshank from California YIMBY has a piece breaking down the strategy for fixing zoning:
Cruickshank lays out NIMBY success stories in cities like Austin and Minneapolis adopting YIMBY-style zoning rules and seeing significant improvements in rental prices. These success stories are representative of a broader recognition – at least among Democratic politicians – that restrictive zoning is a major contributor to the housing emergency.
Repeating these successes in the rest of the country will take a long time, and in the meantime, American tenants are sitting ducks for predatory landlords, With criminal enterprises like Realpage enabling collusive price-fixing for housing and monopoly developers deliberately restricting supplies to keep prices up (a recent Blackrock investor communique gloated over the undersupply of housing as a source of profits for its massive portfolio of rental properties), tenants pay more and more of their paychecks for worse and worse accommodations. They can't wait for the housing emergency to be solved through zoning changes and public housing. They need relief now.
That's where tenants' unions come in, as Ruthy Gourevitch and Tara Raghuveer of the Tenant Union Federation writes in their piece on the tenants across the country who are coordinating rent strikes to protest obscene rent-hikes and dangerous living conditions:
They describe a country where tenants work multiple jobs, send the majority of their take-home pay to their landlords – a quarter of tenants pay 70% of their wages in rent – and live in vermin-filled homes without heat or ventilation:
Public money from Freddie Mae and Fannie Mac flood into the speculative market for multifamily homes, a largely unregulated, subsidized speculative bonanza that lets the wealthy make bets and the poor pay their losses.
In response, tenants unions are popping up all across the country, especially in red state cities like Bozeman, MT and Louisville, KY. They organize for "just cause" evictions that ban landlords from taking their homes away. They seek fair housing voucher distribution practices. They seek to close eviction loopholes like the LA wheeze that lets landlords kick you out following "renovations."
The National Tenant Policy Agenda demands "national rent caps, anti-eviction protections, habitability standards, and antitrust action," measures that would immediately and profoundly improve the lives of millions of American workers:
They caution that it's not enough to merely increase housing supply. Without a strong countervailing force from organized tenants, new housing can be just another source of extraction and speculation for the rich. They say that the Federal Housing Finance Agency – regulator for Fannie and Freddie – could play an active role in ensuring that new housing addresses the needs of people, not corporations.
In the meantime, a tenants' union in KC successfully used a rent strike – where every tenant in a building refuses to pay rent – to get millions in overdue repairs. More strikes are planned across the country.
The American system is in crisis. A country that cannot house its people is a failure. As Rachael Dziaba writes in the final piece for the package, the situation is so bad that water has started to flow uphill: the cities with the most inward migration have the least job growth:
It's not just housing, of course. Americans pay more for health care than anyone else in the rich world and get worse outcomes than anyone else in the rich world. Their monopoly grocers have spiked their food prices. The incoming administration has declared war on public education and seeks to relegate poor children to unsupervised schools where "education" can consist of filling in forms on a Chromebook and learning that the Earth is only 5,000 years old.
A system that can't shelter, feed, educate or care for its people is a failure. People in failed states will vote for anyone who promises to tear the system down. The decision to turn life's necessities over to unregulated, uncaring markets has produced a populace who are so desperate for change, they'll even vote for their own destruction.
