From his latest article:
One of my sources has come forward and brought me a story that will possibly burst the AI bubble. The reason they brought this to me is that I’ve shown — and will continue to show — that I actually give a shit about this industry and the people in it. If you’re wondering what the story is, know that it’s the information I’ve wanted for years, delivered as I have always wanted it, and I will treat it with the reverence it deserves. Imagine what the worst possible thing for me to get would be and you’re probably close.
Link: https://www.wheresyoured.at/ai-is-slowing-down/
I must say I’m pretty excited.
It makes me happy to see that Ed’s finally started his redemption arc after being laughed at or ignored by mainstream media for years.
Bloomberg had him on a few days ago and he didn’t even swear once. :D
Who?
Blogger/podcaster who does a lot of analysis on the financials of AI companies and has called the industry a bubble for a while now. He’s predicted the bubble popping for a while so take the headline with a grain of salt. He’s better at calling out the state of things than predicting what will happen
Host of the Better Offline podcast.
Ed does a huge amount of research creating reverse engineered financials for companies that are doing their best to hide it. As another poster said, he brings the receipts. That’s his main game. Its why you read him in the first place. He backs it up.
What does he have? I’ll bet its inside financial data. I think he got a cache of straight info that lets him correct many of his estimates and prove his worst predictions.
Didn’t OpenAI just submit confidential documents for their IPO? My bet is his source got eyes on it and the information within is damning.
This reminds me of that early episode of Mr. Robot where they need to get the security guard away from his post and they spoof a text from his wife that said something like “come home quick, it’s what we were most afraid of.”
edit: I looked it up and I had it backwards, it was a woman and the text was spoofed to be from the husband saying, “I’m at the hospital. It’s what we always feared.”
I like Ed, but not a fan of this style of teasing. Reminds me of conspiracy theory communities. We’ll see what he has I guess.
I really like Ed but I’m not a fan of his style of anything lol
I wouldn’t be a fan of his style if he was constantly full of it.
Instead, he brings the receipts. Over and over again. So I trust him enough to appreciate the bombastic style. Honestly, you kind of need that to cut through the noise of the actual AI-pilled sloppified b.s. out there.
It’s a bit like thunderfoot who is Elons biggest hater and is basically always right about him, but I also think he’s a sneering little goblin creature that just needs to just shut the fuck up for once.
I can’t watch his videos anymore because it’s like nails on a chalkboard but I’m like this the whole time.
Thundershit is right about Elon accidentally, he knows shit about what he’s talking about, ever, and the fact that his flailing accidentally felt on the right side of this particular issue doesn’t give him much credit.
Yes! My feelings exactly.
I’ve been subscribed for a while, I don’t remember him ever saying something like this so confidently. I do actually trust that he will deliver, and would feel a little betrayed otherwise. He has never hidden important newsworthy leaks behind a paywall, or withheld things longer than a responsible journalist would. We will see soon whatever it is.
I like Ed, but I don’t think there’s anything he could say or do that would blow up the AI grift. It just has too much money riding on it now. I think it will blow up on its own at some point. Until then, the most someone like him can do is to put out stories so that its demise is seen as obvious in retrospect once it happens.
the casino is designed to build and pop bubbles, anyone with the tiniest bit of experience in this rodio knows AI is a shitball.
the issue is…there is loads of $ to be made passing a shitball around, just a matter of hedging properly so your not holding the bag ultimately.
there’s an old Soros quote that goes something like “if you find your in a bubble, your job is to jump in and inflate it as much as possible”
I also doubt it’ll be enough to “blow” the grift. “The market can remain irrational longer than you can remain solvent”. Granted, Altman, Thiel and other scum are already feeling the metaphorical knife edging closer to their throats, because investors are just like loan sharks.
I agree with you. I will say, the only way to blow this up would be to go for the investors. To scare them away. But like you say, the market is irrational. Hell, we see this with SpaceX and Tesla that are both grossly overvalued.
I am still looking forward to his revelations, because I like to watch pro AI gremlins suffer.
I’ll speculate:
- detailed AI usage data from either Azure or AWS
- details about the complete failure of an AI implementation drive from a huge company
- really crazy: detailed financials from either OpenAI or Anthropic
Yeah, first thing that came to mind after reading his statement was that he’s probably got some financial report from one of the big companies
I think it is the last one for OpenAI. Like a draft or thread about their S-1 form
I’ll get excited when it finally blows up. (to be clear, not downvoted because of this. But I’ve just seen bubbles for too long)
Imagine what the worst possible thing for me to get would be and you’re probably close
I mean, that’s probably something like kidney stones, but I really hope I’m not anywhere close for his sake
I think the actual worst thing might be evidence of a widespread unresolvable regression problem in the model design.
If the model training is incapable of distinguishing between AI generated data and human generated data, and unable to identify accurate data from large quantities of inaccurate data, and they’ve known about this the whole time, and they can’t figure out how to keep the models up to date with ongoing information gathering without ingesting data into the models that is toxic to the model development and usefulness…
Well then LLM progression has an expiration date. And all this investment is for naught because in X years you won’t be able to train new models without doing further irreparable damage to functionality and usefulness of the models.
And if AI companies knew about this problem, this pending cliff they were going to fall off, didn’t inform investors, and promised continued growth they knew couldn’t be achieved without a major scientific breakthrough or technology pivot, that would be pretty fucking horrible for the business of LLMs and the economy.
Wow, yeah that seems like something that’s totally plausible, in fact I wouldn’t be surprised at all if you were right.
Exactly. My bet is on “there’s a big unsolvable problem and they all know it’s going to tank the industry growth rate and they never informed anyone, hoping they could solve it in time” in some form. It’s just so on-par with how these companies operate for the most part.
And I say that as a proponent of ML (not “AI” aka “LLMs with skills”)
I suppose that’s what all those BS AI news stories were about, and like moltbook and all that other junk. Just trying to keep it alive for longer to fix it or drum up more money or get the money out of the country idk. Sheesh.
Who knows at this point? There is no shortage of grifters or “true believers” so it’s hard to say. But having worked with several different LLMs across a variety of use cases, I can say for sure that none of them are being completely honest about their shortcomings. Each model has strengths and weaknesses, all of them have unresolved issues, and all of that stuff is being swept under the rug, with the products being touted as one-size-fits-all problem solvers.
Because it’s profit driven, not research driven. The people holding the reigns really don’t care about outcomes for the people using the products, only whether they can get paid for it.
He’s pretty thoroughly covered how using just the available public information explains the insider deals and how demand for LLMs is not enough to justify the cost of developing at the scale that it is. It’s honestly hard to imagine what inside information could possibly worse.
Unless it’s something way out of left field like Nvidia chips are made of freeze-dried orphans or that Jeffrey Epstein invented LLMs.
I bet its evidence that the people involved in this are fully aware of the situation and are making plans for how to cash out and leave us holding the bag when it falls apart.
We can (and should) assume that to be the case, but actual evidence would be pretty significant.
Isn’t that what this is? Taxpayer bailout at the tippy top valuation?
Yeah I’m not quite sure, I guess the actual worst think of is something like they already got caught doing, having actual humans (slave labor) being the ones who are doing all the work in some place far away, not AI, not LLM, nothing. That would be pretty insane I guess.
I have a much more active imagination than you or Ed.
The worst is something like that we’re living in the Matrix and they’re using our brains to do some form of computation as we sleep. If we wake up into the real world, we’re not going to be wearing comfy sweaters, eating porridge, learning kung-fu, and going around in a hover-sub, instead we’re brains connected to bodies that have never had to move, and every minute being awake in that world would be a horror. The connection to AI is that the invention of LLMs and “AI” in the simulation is drawing too much computation costs for the simulation system to manage, so they’re going to reset the matrix to an earlier version soon. But, they’ve discovered we have the most vivid “dreams” if the daily news is horrific, and vivid dreams are good for their needs, so that’s what they’ll be using now to juice the power of their “coppertops”.
Haha, well what seemed like a nightmare scenario in 1999 is not so different from daily life in 2026 as it stands. People are just locked into their own worlds instead of being used as batteries, I suppose they’re technically being used as piggy banks.
The worst (best) thing I can imagine is that they have some sort of mathematical proof that inference costs won’t drop or that training costs will. Their only path to profitability was a moat in the form of high cost to train a model and dirt cheap inference they can resell at a high price. If they finally have some idea which way it will fall then there’s nothing they can do.
I think that this would be too rational and scientific to cause an irrational market to correct.
I’m leaning more towards a money related confirmation. Maybe an OpenAI or Anthropic internal shared numbers.
Or, one of the four horsemen of the AIpocalypse?
There’s one company trying to etch models into the silicon, they got a working prototype in a 8b model and it does 17k t/s. They suggest at data center scales with bigger models it’d pay for itself in a year in saved costs in electricity, cooling, space required etc.
You can apply lora’s to it, but it will fall behind as it ages, im not quite sure how long a static model would last to make up the money after that 1st year, but for something like creative writing where it doesnt need current data (e.g programming but the languages are always changing and evolving is a problem) it might work?
Can’t wait to hear what Ed’s found. Man deserves an award for the way he’s covered this story
Did they tease a date? I listen to them but they have been saying the same kinda of thing about AI for months. (I’m talking about his AI commentary in general, not this latest thing)
they have been saying the same kinda of thing about AI for months
Definitely true, but this grift has been going on for years, someone has to cover it.
I guess I’m just saying, watching ed do interviews, for the last months he has been saying nothing new.
Which is fine, it’s not his job to make up news.
Next two weeks
