I love LLMs, I hate hype

Article URL: https://geohot.github.io//blog/jekyll/update/2026/07/12/i-love-llms.html

Comments URL: https://news.ycombinator.com/item?id=48883343

Points: 366

# Comments: 230

Hacker News 讨论

366 points · 230 comments · 查看原帖

SwellJoe
This line: "this is my main argument against the valuation of frontier labs. It’s not that AI won’t create that much value, it’s that they won’t capture it." 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 t
hamandcheese
> where’s all this new magical software that the productivity improvements should imply? 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.
TheAceOfHearts
At least for me, the jump in productivity has resulted in building stripped down one-off software for my highly specific use-cases. 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.
kenforthewin
I felt the same way in 2024-2025. Then Sonnet 4 was released, and things started feeling different. Opus 4.5 was another step change for me. Everything feels like it's accelerating, and timelines are getting crunched. I guess in some ways I envy OP, who would "bet everything" against ASI - the truth is I don't know, and I don't think anyone knows, where this ends.
andy99
He says he might have been too harsh in his “eternal sloptember” post from may: https://geohot.github.io/blog/jekyll/update/2026/05/24/the-e... I wonder what he thinks was too harsh, still seems pretty bang on, I think it’s going to age well.
dom96
I love LLMs too, but I am concerned about their cost. They are all still very subsidised. Is there any guarantee that I'll be able to run a Opus 4.8-level model on my personal computer before the big AI labs decide to hike up the prices?
LugosFergus
Since no one is talking about it: T2 isn’t about machines taking over the world. That has happened (or will happen). But humans eventually defeat the machines. Skynet is trying to prevent that by killing John Connor. That’s what the movie is about. I suppose it’s also about John searching for a parental figure through the T800. He doesn’t get that through is foster parents and his estranged mother. Anyway, I don’t think this dude actually watched this movie. It’s too bad because it’s a classic.
fwlr
I get it, I want to agree, I really do like the “this is a new tool in the toolkit of the professional software craftsperson” argument… …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.