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

How the AI Bubble Bursts

martinvol.pe

Post reply on HN
Print view View original

431–440 of 557 posts

Re: How the AI Bubble Bursts

#431

Post by tasuki »

logravia wrote: post #350

The thing I am struggling with is where is the impact of LLM tools, especially given the massive increase in token consumption from 2025 to now and the saturated presence of LLMs everywhere. Naively speaking, I have so many expectations for the impact of this tech. I'd expect a noticeable uptick in applications published on Google, Apple and Microsoft app stores. I'd also expect an uptick of games published to Steam.…

> ⸻

Wow, I'm impressed at your usage of this. Apparently it's 0x2E3B, named "three-em dash".

You must be human!

Re: How the AI Bubble Bursts

#432

Post by jsnell »

doom2 wrote: post #429
aurareturn wrote: post #92

This is an awful article. I don't know how it reached #1 on HN. Bottom line is that H100 prices are near 3 year highs, A100s are still profitable to run, B200 prices are increasing, no one has enough compute. Google, OpenAI, Anthropic, Meta, AWS, Azure are all compute constrained. Every single one of them said so publicly. Neo clouds are telling customers they're all sold out now and you even have to book compute in…

> Bottom line is that H100 prices are near 3 year highs, A100s are still profitable to run, B200 prices are increasing, no one has enough compute. Then why aren't the hardware manufacturers of components needed by AI companies making plans yesterday to bring new fabs online to meet demand? That isn't a gotcha question, I genuinely want to know. The money involved isn't that much compared to the money changing hands b…

A new fab will need to be filled with advanced equipment like lithography machines. They are the most complex thing humanity has every built.

There is one supplier of EUV lithography machines in the world, ASML. They are basically acting as an integrator for hundreds of highly specialized components manufactured to unimaginable levels of precision. Each of them has roughly one eligible supplier in the world who are operating at full capacity. To expand, they'll need yet another set of specialized and almost impossible to build equipment.

So the supply chain moves incredibly slowly, and the slowness is intrinsic due to the complexity and depth of the supply chain. It can't be fixed with just money. IIRC ASML is aiming to merely double their production of EUV lithography machines by 2030.

Re: How the AI Bubble Bursts

#433

Post by hintymad »

joshstrange wrote: post #53

> RAM prices are crashing because new models won’t need as much Reality begs to differ [0] and following the link for that text goes to an article [1] where they talk about Google's TurboQuant which supposedly will lower the RAM requirements. Now if that means RAM prices come down (as speculated, not reported on, in the link) or the AI companies just do more things with their extra ram is yet to be determined. The fa…

Some also argue that the RAM price keeps rising because of the bullwhip effect. I was wondering if there's anyway for us to differentiate a sustained demand from the bullwhip effect.

Re: How the AI Bubble Bursts

#434

Post by tracker1 »

Datacenters themselves are really weird... most of the announced 2024 data centers are nowhere near completion, most of NVidia's production is taking longer to deploy than to produce and will be upwards of 2+ years behind on deployments sometime in the next year.

That doesn't even begin to cover the lack of actual electricity to power the data centers. We have more "dark silicon" sitting in boxes that aren't close to being deployed, while a lot of actual people can't manage to buy consumer products for anythign resembling reasonable... it's kind of insane to say the least.

Re: How the AI Bubble Bursts

#435

Post by logravia »

tasuki wrote: post #431
logravia wrote: post #350

The thing I am struggling with is where is the impact of LLM tools, especially given the massive increase in token consumption from 2025 to now and the saturated presence of LLMs everywhere. Naively speaking, I have so many expectations for the impact of this tech. I'd expect a noticeable uptick in applications published on Google, Apple and Microsoft app stores. I'd also expect an uptick of games published to Steam.…

> ⸻ Wow, I'm impressed at your usage of this. Apparently it's 0x2E3B, named "three-em dash". You must be human!

Oh yeah, a month ago I was reading a comment section about LLM writing tendencies and someone humorously suggested using the loooooooong-em-dash to distinguish yourself from LLMs. I found it so charming that I made my keyboard output it when I double tap "-".

On Linux you press Ctrl+Shifs+U and then type 2E3B, then press enter.

Re: How the AI Bubble Bursts

#436

Post by tracker1 »

joshstrange wrote: post #53

> RAM prices are crashing because new models won’t need as much Reality begs to differ [0] and following the link for that text goes to an article [1] where they talk about Google's TurboQuant which supposedly will lower the RAM requirements. Now if that means RAM prices come down (as speculated, not reported on, in the link) or the AI companies just do more things with their extra ram is yet to be determined. The fa…

Even worse, 3 memory companies control well over 90% of the international market, with a history of cartel collaboration that's going to be ever harder to prove with fewer companies.

Re: How the AI Bubble Bursts

#437

Post by aurareturn »

tartoran wrote: post #392
fsloth wrote: post #293

Earlier quoted context omitted.

I'm paying $20 for Codex and $90 for the Claude Max plan. They are a "pry from my cold dead fingers" product for me. IMO if someone tried this tech last time 6 months ago, or their only exposure is eg. via MS copilot, they do have a rational reason for skepticism. No technology of this complexity has improved this rapidly in my memory (well, ok, we had the CPU speed races from 90's to early 2000's).

Would you still pay if prices were to increase,say $1500-2000 monthly?

Probably. I assume the value would drastically increase. Companies will definitely continue to pay for it. It's irreplaceable now.

Re: How the AI Bubble Bursts

#438

Post by aurareturn »

doom2 wrote: post #429
aurareturn wrote: post #92

This is an awful article. I don't know how it reached #1 on HN. Bottom line is that H100 prices are near 3 year highs, A100s are still profitable to run, B200 prices are increasing, no one has enough compute. Google, OpenAI, Anthropic, Meta, AWS, Azure are all compute constrained. Every single one of them said so publicly. Neo clouds are telling customers they're all sold out now and you even have to book compute in…

> Bottom line is that H100 prices are near 3 year highs, A100s are still profitable to run, B200 prices are increasing, no one has enough compute. Then why aren't the hardware manufacturers of components needed by AI companies making plans yesterday to bring new fabs online to meet demand? That isn't a gotcha question, I genuinely want to know. The money involved isn't that much compared to the money changing hands b…

They are. They're making as many fabs as they can as fast as they can.

The bottleneck is ASML, who can only make so many EUV machines. No one else can make EUV machines.

Scaling chip fabs and chip equipment is much harder. And you have to understand that chip fabs go bankrupt if demand suddenly drops so they have to be more cautious by default.

Re: How the AI Bubble Bursts

#439

Post by aurareturn »

nunez wrote: post #372
aurareturn wrote: post #92

This is an awful article. I don't know how it reached #1 on HN. Bottom line is that H100 prices are near 3 year highs, A100s are still profitable to run, B200 prices are increasing, no one has enough compute. Google, OpenAI, Anthropic, Meta, AWS, Azure are all compute constrained. Every single one of them said so publicly. Neo clouds are telling customers they're all sold out now and you even have to book compute in…

> I think AI agents could completely replace Microsoft Office How? What do you think lawyers/government will use to write briefs?

With ChatGPT

Re: How the AI Bubble Bursts

#440

Post by paulddraper »

drakythe wrote: post #195
paulddraper wrote: post #87

Earlier quoted context omitted.

Anthropic has said inference is profitable. That’s a biased source, but the math pencils. This is why switching to local open weight models saves a lot of money. (Even though it’s not apples to apples.)

Anthropic also recently tweaked their usage limits to discourage use during peak hours. Why would they do that if inference was profitable?

Those are subscription plans. They tweaked the limits/periods included in the subscription. Having higher limits for subscription plans didn't give them any more revenue.
Post reply on HN