Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

I love LLMs, I hate hype

geohot.github.io

Post reply on HN
Print view View original

171–180 of 340 posts

Re: I love LLMs, I hate hype

#171

Post by fragmede »

anuramat wrote: post #157
maxgashkov wrote: post #144

Earlier quoted context omitted.

If they haven't relaxed the classifier this makes little to no difference as the model is essentially unusable no matter the token quotas.

what are you working on? I only hit the guardrails twice after burning through two weeks of 20x max plan, both times on ML stuff; still more than I'd want to, but not unusable

Lucky for you. I, along with everyone else on the fable guardrail thread have been hitting guardrails with Fable for boring normal shit and getting downgraded.

Re: I love LLMs, I hate hype

#172

Post by intothemild »

SwellJoe wrote: post #81

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, thoug…

> 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.

Are they really the best models? Like take anthropic. Without mythos, it's the what? Third best?

Sure openAI just leapfrogged them but .. seriously to get there it's a giant model that costs insane per token.

Nobody needs that, it's like NVIDIA or Intel claiming they have the best gaming performance, but to achieve that they are using more power per frame than anything else.

Re: I love LLMs, I hate hype

#173

Post by mike741 »

InvertedRhodium wrote: post #126
dakolli wrote: post #104

Earlier quoted context omitted.

The fact that you're even saying this it is probably an admission that you do think it's making you dumber. Most people I know, who are honest with themselves, have admitted to me that they feel like it's making them dumber or "zombifying" them. This is also well studied already, https://arxiv.org/abs/2506.08872 LLMs are poison for the brain, I'm almost certain of it, at least when used in the way most people are usi…

Socrates thought the same of reading and writing, that it would weaken the memory and isolate people from one another. 1966 saw the peak of calculator protests, where math teachers claimed similar things of calculators.

all of those predictions seem to have come true

Re: I love LLMs, I hate hype

#174

Post by int_19h »

lacedeconstruct wrote: post #122
dolebirchwood wrote: post #118

Earlier quoted context omitted.

> This doesn't make sense Makes perfect sense to anyone good at using these models. What doesn't make sense is that analogy. Typing prompts isn't even close to as difficult to baking bread.

>Makes perfect sense to anyone good at using these models. It doesn't really, because whenever I ask them what did they actually create, its always a shitty dashboard or a finance tracker or something derivative and worse than what is out there

I'd say that's more indicative of what kind of things people in your bubble generally work on.

I vibe-coded a semantic parser for Lojban.

A friend of mine is using it to work on dev tooling.

Another friend, a mathematician, has recently used it to prove a conjecture he published 15 years ago.

Re: I love LLMs, I hate hype

#175

Post by slopinthebag »

I think I agree completely. It’s worth pointing out that Linus’s comparison of LLMs and compilers is that they are both tools, not that they’re the same thing. Like how both compilers and hammers are tools, but a hammer is not a compiler.

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?

Re: I love LLMs, I hate hype

#176

Post by int_19h »

dom96 wrote: post #39

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?

We're supposedly getting Mac Studio with 1.5Tb RAM in 2 years. That would be enough to run an Opus-level model.

Of course, it will also probably cost somewhere around $50k...

But if local AI really does become pervasive, maybe it'll be one of the things people buy on credit, like cars.

Re: I love LLMs, I hate hype

#177

Post by slopinthebag »

gyomu wrote: post #137
SwellJoe wrote: post #81

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, thoug…

In 5-10 years an Apple Watch will run a Fable level model locally. I don’t think we (hackers) should worry too much about token cost inflation. The current wave of providers, that’s another story.

You can’t seriously think an Apple Watch will have hundreds of gigs of vram in 10 years?

Re: I love LLMs, I hate hype

#178

Post by int_19h »

kenforthewin wrote: post #92
lukeschlather wrote: post #57

Earlier quoted context omitted.

He didn't say he bets everything against ASI, he said he bets everything against ASI being a flash of light in the sky which destroys our chance of getting access to the wealth it creates.

That's a much less generous interpretation of his writing. "Yes we will birth superintelligence, but everything will just sort of work out for us humans". This seems like a silly take to me.

Is it? To me, the notion that a superintelligence (by which I'm assuming you mean the more sensible "something smarter than us", not "literally a godlike entity") automatically means that sky is going to fall is sillier.

Re: I love LLMs, I hate hype

#179

Post by int_19h »

reinitctxoffset wrote: post #90
password54321 wrote: post #20

Yeah I don't think any of the labs have some secret sauce for intelligence either. It seems most of the advancements are still coming from hardware, making LLMs more efficient and throwing more compute and data at problems. And even those problems still require a lot of prompt engineering: https://cdn.openai.com/pdf/04d1d1e4-bc75-476a-97cf-49055cd98...

I'm pretty sure at this point that Anthropic is training mixture models (at least in the heavy pre-train) and deploying them dense with explicit loss on thinking trace coherence. Having a thinking trace that is legible, coherent, and immediately implies the explicit turn output and/or tool use seems difficult if not impossible to reliably get from mixture models. I predict MoE is a transitional technology, it's got t…

MoE is just activating fewer weights per token than the whole model. It will continue to make sense for as long as compute is more expensive than memory (at scale).

Re: I love LLMs, I hate hype

#180

Post by fragmede »

simoncion wrote: post #140
wizzwizz4 wrote: post #85

Earlier quoted context omitted.

But the part of the ear that needs cleaning can be reached without a cotton bud. This is like shoving a sponge down your windpipe to remove mucus.

> This is like shoving a sponge down your windpipe to remove mucus. In my personal experience, not using a cotton-tipped swab for the task is like cleaning a plate loaded with gunk and burned-on patches with one's bare hands rather than choosing to use a sponge and/or brush. You can do it, [0] but it's much more work, much more time consuming, or you get an inferior result. [0] In my case, I'd need to make one set of…

We live in the future. You can get an ear cleaning camera endoscope device for $40 next day Amazon delivery anywhere they reach.
Post reply on HN