Memory prices have doubled, Macs and iPads have gone up, and iPhones are expected to follow. Ed Zitron – who writes the Where's Your Ed At newsletter, hosts the Better Offline podcast, and has been described by Politico as the AI boom's most "acerbic gadfly" – has spent years arguing the buildout driving those costs will never pay for itself.
We asked him what happens to Apple if he's right.
You've been calling AI a bubble since before it was fashionable. For MacRumors readers who mostly know it as ChatGPT or Apple Intelligence on their iPhone, what, in plain terms, is actually broken about the economics of the LLM industry?
At their very core, Large Language Models' costs run contrary to basically every model of selling software.
Consumers and enterprises alike have been trained to pay a monthly fee for a service, and while these services might have limits or strictures, basically nobody buying software expects to have a metered service, let alone one that's both metered and with hard to measure costs.
LLMs burn tokens at a per-million rate regardless of whether or not you get the response you want, or whether it does what you ask it to do. If you ask a coding agent to do some sort of software task and it goes off and spins its wheels in a loop, you're paying for the tokens regardless.
AI companies knew that consumers would never pay the actual cost of their AI services, so they have, for the most part, sold them monthly subscriptions with vague rate limits that allow them to burn way more in tokens than the cost of their subscription. SemiAnalysis found that you can burn hundreds of dollars on a $20-a-month subscription and thousands of dollars on a $200-a-month subscription, and while AI boosters will claim that these companies have "70% gross margins on tokens," there is little proof that this is the case, and my own reporting shows that OpenAI lost $20.9 billion on $13.07 billion in revenue in 2025.
This means the very basic economics are broken. If Anthropic and OpenAI believed customers would actually pay the real cost of AI tokens, they wouldn't have to give away 20 to 40 times the amount of tokens to subscribers.
Meanwhile, back in March of this year, both moved their enterprise customers over to token-based billing. Within a few weeks, it came out that Uber had spent its entire annual token budget in the space of a quarter, and its COO said that it was getting "harder to justify" the cost of AI because it was hard to track the cost of AI to any actual useful features shipping. Sam Altman would eventually say it was a "huge issue" but declined to say how it might be fixed.
This is a problem across basically every single AI-powered startup, which has to pay the per-million token rate. Perplexity, Cursor, GitHub Copilot (which moved to token-based billing in June) – every single AI startup is unprofitable because their users don't want to pay the actual cost of AI.
Another issue is that AI services are just not that useful or differentiated. While people get some sort of benefit out of AI-generated code, these tools actually end up making them slower, and are filling codebases full of slop. Otherwise, an LLM is an LLM is an LLM – it can generate, it can summarize, it can search, and that's about it, which means that every AI service is effectively the same. That's why 89% of all AI revenues are Anthropic and OpenAI, and why every AI startup talks in terms of "annualized revenue" (monthx12) – because actual revenues are very depressing. Even then, most are barely at $100 million annualized.
Then there are the data centers. An AI data center is very, very expensive to build, takes 18 to 36 months, and costs billions of dollars, which means effectively anyone building one will be raising debt and only get paid once a customer moves in... except there aren't really any customers for AI data centers outside of Anthropic and OpenAI, both of whom are so unprofitable that they've had to raise hundreds of billions of dollars even when Microsoft, Google and Amazon built all their infrastructure.
The only reason everybody isn't freaking out about this is because AI-related stocks have done well, even though none of the hyperscalers actually share their AI revenues.
You've argued that AI's demand story is essentially a mirage – that most of the data center capacity is being absorbed by OpenAI and Anthropic themselves, which is masking the absence of real enterprise demand. If that's right, who do you think will actually bear the cost when the whole thing unravels?
Honestly, it's going to be a lot of private credit funds, because they're the ones funding the data centers, and they're funded by pension funds like the SF teachers fund or CalPERS, which makes me really, really worried about the systemic contagion.
People will argue that this means there's going to be a bailout, but this isn't really a bailoutable thing. These data centers are funded by project financing, which means that the money is basically gone and the only way to "make them whole" would be to either buy out the debt or feed them revenues. While you could theoretically bail out these special purpose vehicles (SPVs), doing so would be to the tune of hundreds of billions of dollars and be political cancer.
I also fundamentally believe that Oracle gets killed by OpenAI. Its revenues have been stagnating for 20 years, and the only way it's kept its head above water is $85bn+ in acquisitions, and even then, that's just kept things flat. Its bets on AI data centers – $340bn+ with hundreds of billions in debt – require OpenAI to become the largest, most-profitable company in the world by 2030, or Oracle runs out of money. Good luck on that one Larry.
The actual contagion from the collapse will be widespread. The TWSE is heavily reliant on Taiwanese ODMs like Quanta and Hon Hai (Foxconn) who have seen their revenues boosted by selling AI servers. While I don't think it kills these companies (especially as Hon Hai makes so much from Apple), it will cut their stock price, which will in turn cut the livelihoods of those investing. The same goes for Korean investors on the KOSPI, and eventually American investors in the various hyperscalers and semiconductor companies that are running entirely because of the AI bubble.
It all really sucks.
Hyperscalers building out data centers have hoovered up so much of the world's memory supply that DRAM prices have roughly doubled this year. Tim Cook has called Apple's recent price increases "unavoidable" – Macs and iPads are already up, and iPhone models are expected to follow soon. It seems the clearest sign yet that the bubble's costs are being passed onto ordinary consumers. If the buildout is as speculative as you say, are people basically paying more for hardware to subsidise data centers that may never turn a profit?
I don't think any of them turn a profit, no.
Hyperscalers have spent over $1 trillion in capex since 2022. If we assume that even half of that is AI data centers, they need to make over $1.5 trillion in brand spanking new profit, not revenue, but actual profit, to justify any of these expenditures.
And yes, we are all paying more for stuff for effectively no reason other than that everybody in big tech has gone insane and wants to build as many data centers as possible. I wrote about this in the Hater's Guide To The Memory Crisis.
The hyperscalers are collectively spending north of $650 billion on AI infrastructure this year, whereas Apple is spending about $14 billion. It's as if Cupertino is treating AI as a commodity by paying Google around a billion a year for Gemini to run Siri and doing what it can on-device. The company was previously said to be dangerously behind on AI. That no longer seems to be the case, but at the same time, is it really more savvy to be renting a dependency on the very companies you say can't survive?
So, I think Apple Intelligence was both the worst and best thing to happen to Apple as far as the AI bubble goes.
It was a mass-radicalization of its users against AI – an attempt to cram a barely-functional series of add-ons that nobody asked for in the most-obtrusive way possible, with summaries that were almost immediately turned into memes and a new Siri that was, somehow, even worse than the old Siri.
And I think that told Apple to pump the brakes. It's barely spent anything on capex. It's barely done anything with AI. Despite headline after headline claiming it's "falling behind," nobody can really explain what it is it's falling behind on or why it matters. People hate Apple Intelligence, and I think Apple knows that, and so they're going to jingle the keys for the markets by putting "AI" on stuff without ever really putting their back into it.
If the bubble deflates the way you expect (write-downs, too much compute supply, possibly OpenAI cratering), can you walk us through what that would actually look like for Apple? Does anything break for iPhone users, or does Apple mostly watch it happen from the sidelines? Could it even benefit Apple?
I think things would look much the same for Apple. I think they will sit on the sidelines and watch everything burn. I could see them doing some choice acquisitions as things begin to collapse, but I could also see them do nothing.
I think Apple is in a very weird place at the moment. The Vision Pro was a dud, but it was also the most interesting and future-forward thing I've seen anyone put out in a while. If they were smart, they'd tread water and sink as much money into making that as small as humanely possible – no matter how long it takes – because the entire AI bubble is a result of everybody running out of hypergrowth ideas, mostly because we're flat out of new interfaces.
I want to be clear that what they want to do with the Vision Pro requires it to basically be weightless and invisible and never need adjustments. When it works, it's genuinely awesome. But that's a load-bearing when. One slight movement means the whole thing goes out of focus. I can't even use mine anymore because it needs an update that requires you to wear the thing the whole time. So much promise, released too early, shoved out the door by a CEO on his way out.
