Our Big Beautiful AI Bubble(s), ChatGPT-5, Capitalism, China, ASOIAF fan fiction, and salsa
Tech Bubble Consumer Dispatch #6: What I’ve been reading, watching, and listening to (8/26/25).
Welcome back, valued Tech Bubble Consumer. This is for paid subscribers, where I'll talk a bit about what I've been reading and watching and listening to. If you'd like to get my recommendations directly jacked into your brain, consider subscribing for $7 a month (the cost of a box of Kenyan tea my father enjoys) or $70 a year (a few ill-advised drinks at the last reading you were dragged to).
It’s been a minute so this week’s round-up will run long: it includes essays on our ongoing AI bubble, GPT-5, debates about the impact of digital technologies on capitalism, China, a movie/show recommendation, books, and music I’ve enjoyed, and more.
Table of Contents
AI Bubble
GPT-5
Capitalism
China
Movies/Shows
Music
Books
AI Bubble
How to Argue With An A.I. Booster - Ed Zitron, Where’s Your Ed At
Zitron is easily one of my favorite commentators/analysts/writers on the AI bubble's financial and consumer dimensions. I recommend you read every single essay he's published, but I especially recommend his latest. The spirit of it:
In the last two years I've written no less than 500,000 words, with many of them dedicated to breaking both existent and previous myths about the state of technology and the tech industry itself. While I feel no resentment — I really enjoy writing, and feel privileged to be able to write about this and make money doing so — I do feel that there is a massive double standard between those perceived as "skeptics" and "optimists."
To be skeptical of AI is to commit yourself to near-constant demands to prove yourself, and endless nags of "but what about?" with each one — no matter how small — presented as a fact that defeats any points you may have. Conversely, being an "optimist" allows you to take things like AI 2027 — which I will fucking get to — seriously to the point that you can write an entire feature about fan fiction in the New York Times and nobody will bat an eyelid.
This reminds me of an interview with Noam Chomsky where he defines “concision” in terms of our ad-centric media system:
Concision means you have to be able to say things between two commercials. Now that’s a structural property of our media—a very important structural property which imposes conformism in a very deep way, because if you have to meet the condition of concision, you can only either repeat conventional platitudes or else sound like you are from Neptune That is, if you say anything that’s not conventional, it’s going to sound very strange. For example, if I get up on television and say, “The Soviet invasion of Afghanistan is a horror,” that meets the condition of concision. I don’t have to back it up with any evidence; everyone believes it already so therefore it’s straightforward and now comes the commercial. Suppose I get up in the same two minutes and say, “The U.S. invasion of South Vietnam is a horror.” Well, people are very surprised. They never knew there was a U.S. invasion of South Vietnam, so how could it be a horror? They heard of something called the U.S. “defense” of South Vietnam, and maybe that it was wrong, but they never heard anybody talk about the U.S. “invasion” of South Vietnam. So, therefore, they have a right to ask what I’m talking about. Copy editors will ask me when I try to sneak something like this into an article what I mean. They’ll say, “I don t remember any such event.” They have a right to ask what I mean. This structural requirement of concision that’s imposed by our media disallows the possibility of explanation; in fact, that’s its propaganda function It means that you can repeat conventional platitudes, but you can’t say anything out of the ordinary without sounding as if you’re from Neptune, a wacko, because to explain what you meant—and people have a right to ask if it’s an unconventional thought—would take a little bit of time.
To that point, Zitron's work (this essay is 16,000 words long) really hinges on the idea that if you are arguing with an AI booster, you must go long and you must go deep. This is not as high of a hurdle to climb as it sounds. A great deal of commentators are incredibly lazy, incurious, deceptive, or flat-out delusional—Zitron’s work is a breath of fresh air here, as are many of the journalists he directly cites within it. This sort of argument guide covers a wide range of arguments you might encounter when dealing with boosters: comparisons to the Uber/Amazon narrative about growing into profitability, the importance of overbuilding compute infrastructure, AI 2027, the potential of agentic workflows, unit economics of inference, edge use cases that are genuinely helpful or useful, the democratization of X with the introduction of AI vibe coding and co-creation, and more. I’ll pull from the conclusion to pitch it once more
I also think the media is failing on a very basic level to realize that their fear of missing out or seeming stupid is being used against them. If you don't understand something, it's likely because the person you're reading or hearing it from doesn't either. If a company makes a promise and you don't understand how they'd deliver on it, it's their job to explain how, and your job to suggest it isn't plausible in clear and defined language.
This has gone beyond simple "objectivity" into the realm of an outright failure of journalism. I have never seen more misinformation about the capabilities of a product in my entire career, and it's largely peddled by reporters who either don't know or have no interest in knowing what's actually possible, in part because all of their peers are saying the same nonsense.
As things begin to collapse — and they sure look like they're collapsing, but I am not making any wild claims about "the bubble bursting" quite yet — it will look increasingly more deranged to bluntly publish everything that these companies say.