When some huge and stupid public chatbot disaster hits the news, the AI pumpers will Kramer into the mentions to say stuff like “you have to admit, AI is here to stay.”
Well, no, I don’t. Not unless you say what you actually mean. What’s the claim you’re making? Herpes is here to stay too, but you probably wouldn’t brag about it.
What they’re really saying is “give in and do what I tell you.” They’re saying AI like it is in the bubble is a permanent force that will reshape society in Sam Altman’s image. It’s a new paradigm! So you have to give in to it. And give me everything I’m demanding.
Here’s an egregious example from the Washington State school system. It starts with the sentence “AI is here to stay,” then there’s a list of AI stuff to force on the kids, on the assumption that all of this will work forever just like the biggest hype in the bubble. And that’s not true! [OSPI]
If you ask these guys why AI is here to stay, they’ll just recite promotional talking points. So ask them some really pointy questions about the details.
Remember that a lot of these people were super convinced by just one really impressive demo that blew their minds. We have computers you can just talk to naturally now and have a conversation! That’s legit amazing, actually! The whole field of natural language processing is 80% solved!
The other 20% is where the computer is a lying idiot — and it probably can’t be fixed. That’s a bit of a problem in practice. Generative AI is all like that — it’s genuinely impressive demos with unfixable problems.
Sometimes they’ll claim chatbots are forever because machine learning works for X-ray scans. If they say that, they don’t know enough about the details to make a coherent claim, and you’d have to teach them the difference between medical machine learning systems and chatbots before they could.
Grifters will try to use gotchas. Photoshop has AI in it, so you should let me post image slop to your forum! Office 365 has AI in it, so if you use Word then you’re using AI! Spell check’s a kind of AI! These are all real examples. These guys are lying weasels and the correct answer is “go away”. Or maybe something stronger.
Are they saying the technology will surely get better because all technology just improves? Will the hallucinating stop? Then they need evidence of that, because it sure looks like the tech of generative AI is stuck at the top of its S-curve at 80% useful and has not made any major breakthroughs in a couple of years or more.
The guy’s probably seen an impressive demo, but he’s going to have to bring proper evidence that chatbots are going to make it to being any sort of reliable product. And we have no reason to think they will.
Are they saying that OpenAI and its friends, all setting money on fire, will be around forever? Ha, no. That is not economically possible. OpenAI alone needs tens of billions of fresh dollars every year. Look through Ed Zitron’s numbers if you think numbers will do any good to reply to this one. [Ed Zitron]
The big venture-funded AI companies are machines for taking money from venture capitalists and setting it on fire. The chatbots are just the excuse for them to do that. The companies are simply not sustainable standalone businesses.
Maybe after the collapse, there’ll be a company that buys the name “OpenAI” and dances around wearing its skin. The name “Fyre Festival” just went on sale. [eBay]
Are they saying there’s a market for generative AI, so it’ll surely keep going when the bubble pops? There may well be some market — the vibe coders are addicts. But the prices will be at least five or ten times what they are now if the chatbot has to pay its way as a standalone business.
But chatbots are useful to me personally! Sure, they do some useful things. Large language models are based on transformers, so anything a transformer does well, a chatbot will do okay if it’s trained. Translation, transcription, grammar checking, a chatbot can at least muddle through. And right now, the chatbot is convenient. Will you pay ten times the price for that? I’m not so sure.
Are they saying you can always run a local model at home? Sure you can, and about 0.0% of chatbot users do that. In 2025, the home models are painfully slow toys for nerd enthusiasts, even on a high-end box. No normal people are going to do this to get what they get from a casual chatbot now.
I’ve seen the people saying “AI is here to stay” get called on it and back down to, well, the technology will still exist. Sure, mathematics is here to stay. The transformer architecture is actually useful for stuff. But just existing isn’t much of a claim either. Technologies have their heyday then the last dregs of them linger forever.
Crypto is still around, serving the important “crime is legal” market, but nothing else is happening, and it’s radioactive for normal people. If you search for “AI is here to stay” on Twitter, you’ll see the guys who still have Bored Ape NFT icons.
Generative AI has a good chance of becoming as radioactive to the general public as crypto is. They’ll have to start calling the stuff that works “machine learning” again.
So. If someone says “AI is here to stay,” nail them down on what the precise claim is they’re making. Details. Numbers. What do you mean by being here? What would failure mean? Get them to make their claim properly.
I mean, they won’t answer. They never answer. They never had a claim in mind. They were just making promotional mouth noises.
I’ll make a prediction for you, to give an example:
The AI bubble will last at least two, maybe three more years, because the venture capitalists really need it to. When, not if, the VCs and their money pipeline go home and the chatbot prices multiply by ten or more, the market for generative AI will collapse.
There will be some small providers left. Gen-AI will technically be not dead yet! But the bubble will be extremely over. The number of people running an LLM at home will still be negligible.
It’s possible there will be something left after the bubble pops. AI boosters like saying it’s just like the dot-com bubble!! But i‘ve never really been convinced by the argument “Amazon lost money for years, so if OpenAI just sets enough money on fire then it must be Amazon.” It’s not a good argument.
Will inference costs — the real cost of each query, which are 80%-90% of compute load — come down? Sure, they’ll come down at some point. Will it be soon enough? Well, Nvidia’s Blackwell hasn’t been a good chip generation, so Nvidia is putting out more of their old generation chips while they try to get Blackwell production volumes up. So more efficient chips won’t fill out the market very soon.
So there you go. I might be wrong about any of that — but at least I’ve given reasons for what I’m saying.
If you want to say “but AI is here to stay!” then tell us what you mean in detail. Stick your neck out. Give your reasons. You might be wrong about parts of it, but at least you’ll have made a checkable claim.