That is a very astute and concise way to explain everything about how the frontier labs are behaving and how they're trying to push more people to pay token rates for the best models. At the current subscription prices ($100 or $200 a month for a generous, though bounded, amount of tokens), frontier models are a no-brainer, most folks and companies will use them. But, at token rates, 10x or 100x the cost of open models or what I was spending on the frontier models a month ago? That is a harder question to answer "yes" to. I certainly wouldn't spend $1000 a month for the best model, much less $10,000; my employer might pay $1000/month, but definitely not $10,000. The frontier labs need everyone to answer "yes" to spending 100x what they currently spend to justify the valuations, and it's just not going to happen as long as everyone knows how to make these models.
Both OpenAI and Anthropic are trying to figure that out now. Anthropic, in particular, has their finger on the trigger...they want to push people to usage-based billing for Fable. But, OpenAI released 5.6 Sol, competitive with Fable (or close enough), and it's available via subscription (even the $20 subscription!), and there's no moat keeping someone from switching. If Anthropic really does end Fable access on the subscription plans in a few days, I predict a large market move back toward OpenAI.
The market isn't going to bear the cost of making the frontiers investment make sense.
I think an interesting question is going to be, if models are a commodity, who is going to want to foot the very expensive bill to train them? I'm sure training cost will drop.. eventually, but I doubt it will happen fast enough for any of these companies.
It would have been an interesting experiment to charge more for it right away and see what the market would bear, rather than tease it for long enough for it to be presumably superseded any time now by whatever is next.
By the same measure, NVDA is Cisco, providing the backbone and capturing a ton of the early benefits, but soon becomes furniture while the excitement moves further up the chain.
And we can't ignore the power of "good enough". GLM5.2 may not be as good as the SOTA models, but it can be good enough for most, of not all, of our needs.
Think airlines - both passenger and freight. They have never come close to capturing all the economic value they enable.
It's running, privately, in my homelab.
I think we are entering what I call the "have it your way" era. If an open source project doesn't do exactly what you want it to do, fork it, or create a new version. It's too easy.
This makes me a bit concerned about the future of open source. Upstreaming used to be worth it, since maintaining a fork is effort too. But now the balance has shifted significantly. Especially with many projects becoming a lot stricter about contributing, and some becoming outright hostile to AI. I can't blame them. But I think the effect will be that improvements are less likely to make it back to the community as AI adoption increases.
You can use an LLM to create anything but you still need to know what it is that you're building, and you need to think through how everything should work or the LLM will just fill it with sausage. You can tell that the models are still quite jagged and limited by the mixed quality from a lot of the software that these presumed trillion dollar companies are putting out. The future is sausage.
Anyway, I don’t think this dude actually watched this movie. It’s too bad because it’s a classic.
In the past when I couldn't figure out something, I'd take a break for a couple days, while going through Google → Stack Overflow → Reddit, and by the time you got to that point you rarely got useful answers, usually either trolls or silence.
Now I can just ask AI about fleeting ideas and always have a starting point for some area of some project to work on.
A lot/some of the concerns about the AI Age could be alleviated if people got UBI and a 4-day workweek.
like if AI's supposed to be so great why do we still have to work so much??
and if we don't have to work, how do we pay for food and bed?
I wonder what he thinks was too harsh, still seems pretty bang on, I think it’s going to age well.
I do alt inference prototypes and got much farther than I had hoped to. So indeed, any investor in AI should read deep and question hype and frontier lab investments.
See: https://github.com/guilt/TinyToT for the sort of hype busting I do.
…but consider: the Q-tip. “Don’t use it to clean your ears”, but for most people that’s all they want to do with it, and empirical observation indicates that this dynamic results in either “using Q-tips irresponsibly” or “not using Q-tips”, with “uses Q-tips properly” being a small-to-vanishing proportion of the whole.
It's possible to use LLMs without logging onto twitter to be exposed to the people spouting off about a "perpetual underclass." I love the internet, but it really feels like (now more than ever) you have to be intentional about what sites you visit.
Yes, it is a vastly efficient search technique using fast computers.
Wait, does this mean I'm better at something than geohot? All that time spent learning regexps wasn't a waste!
They’re really quite useful but the Bay Area mentality and hype is completely disgusting and turned me completely off for a while. What brought me back was a surge in useful Chinese models, with a significantly more mature approach to marketing and discourse. I think Geohot is 100% correct about SF and the people there perpetuating insanity. And I wonder, has it always been like that there, or is this a new phenomenon?
> And two, this strawman jump from, oh hey, it’s a fancy autocomplete, smart compiler, better search engine, to it’s gonna like own the whole light cone bro like if you aren’t in SF and at the right parties there’s gonna be like a flash of light in the sky one day and you’re not even gonna know what happened but everything just Changed.
Haha, OP has a way with words.
In a way, both these emotional extremes (FOMO & the singularity) are just tools being used to continue driving the massive CapEx behind LLM improvement. Hate to love it? Love to hate it?
So far, all we have is more software running on computers. It's powerful, and it's amazing, but it's not magic.
Calling it "AI" was possibly a net-negative but we don't know yet.
It's bullshit in the sense that they don't know for sure, but the author doesn't either. Why might or might not it be true?
This one is a well deserved read. and I find myself agreeing with @geohot
Even if the blog's title is "Singularity is nearer"
Part-time vibe coder here. As far as I can tell it's no longer "slop" in the sense that I am no longer hitting the state where AI can't maintain what it has written or can't meet the requirements. But shipping is as hard as ever. Projects get more ambitious, scope creeps, nuances keep being discovered as you work on a piece of software, etc. Right now I feel that the absence of new software explosion is best explained by the fact that the part we have automated turned out to be relatively small.
> AI is something that’s happening mostly due to Moore’s law and general progress in computing, not something that they are doing.
But if these companies control the vast majority of compute power, which seems like the plan they are already executing, won't they capture most of the value from the progress of AI?
"Where’s all this new magical software that the productivity improvements should imply?"
This is a recurring gotcha in the anti-AI marketplace of denialism. It's a bit like saying "I saw a fat guy, so why do people keep telling me that GLP-1s change everything?"
It takes time. Like already I would say just about every programmer has replaced a number of tools with random shit they spit out from LLMs. It percolates out from there as some things become products, etc.
And to anyone actually paying attention, and not just feeding their delusions, the impact is utterly enormous. Incontestable. The "Where's the software? / Where's the change?" people are absolutely going to find themselves in the dustbin of history.
It's also fascinating how often people do the stochastic parrot horseshit.
The other night I had to do a large scale compression test with libjxl, which notably is software that has seen an enormous amount of optimization interest and you would assume has little extra to be eked out. I've traced through this software before and the compression path is insanely complex. It's the sort of software that is headache inducing. Anyways, curious what the state of the platform was I grabbed the latest source and asked Fable to look for low-hanging fruit in the lossy and lossless compression paths. It suggested a few, created a test harness to A:B bitwise compare with the original, and implemented its optimizations. It achieved a 14% performance increase in a single pass, using just the remaining quota I had on a subscription as my week drew to a close. And all it did was some high level logical optimizations, some more efficient memory allocations, and so on. All of its code was completely in the style of the project, was no more significant than necessary, and so on. Anyone that isn't utterly blown away by that -- who gets the hype -- is lying to themselves.
It’s why con artists, scammers always flood every hype cycle. Greed ruins everything.
The blog has a tagline, "the singularity is nearer". I think belief in a "singularity" almost implies these things to some degree.