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

How the AI Bubble Will Pop

derekthompson.org

Post reply on HN
Print view View original

261–270 of 531 posts

Re: How the AI Bubble Will Pop

#261

Post by kalap_ur »

I just heard a thesis that there is no bubble unless there is debt in it. Currently mostly internal funds were used for increasing capex. More recently started we seeing circularity (NVDA -> OpenAI -> MSFT -> NVDA), thus this is less relevant so far yet. Especially as around ~70% of data center is viewed to be GPU, so NVDA putting down $100B, that essentially funds "only" $140B of data center capex.

META is spending 45% of their _sales_ of capex. So I wonder when are they going to up their game with a little debt sprinkled on.

Re: How the AI Bubble Will Pop

#262

Post by Gerardo1 »

UrineSqueegee wrote: post #18
piker wrote: post #15

Earlier quoted context omitted.

> Try building something new in claude code (or codex etc) using a programming language you have not used before. Your opinion might change drastically. Try changing something old in claude code (or codex etc) using a programming language you have used before. Your opinion might change drastically.

this is literally how i maintain the code at my current position, if i didn't have copilot+ i would be cooked

...what were you doing before?

That's bread and butter development work.

Re: How the AI Bubble Will Pop

#263

Post by data-ottawa »

maeln wrote: post #142
cornholio wrote: post #64

Earlier quoted context omitted.

It's silly to say that the only objective that will vindicate AI investments is AGI. Current batch of deep learning models are fundamentally a technology for labor automation. This is immensely useful in itself, without the need to do AGI. The Sora2 capabilities are absolutely wild (see a great example here of what non-professional users are already able to create with it: https://www.youtube.com/watch?v=HXp8_w3XzgU…

> Current batch of deep learning models are fundamentally a technology for labor automation. This is immensely useful in itself, without the need to do AGI. The Sora2 capabilities are absolutely wild (see a great example here of what non-professional users are already able to create with it: https://www.youtube.com/watch?v=HXp8_w3XzgU ) > So only looking at video capabilities, or at coding capabilities, it's already…

This is very surface level criticism.

Qwen image and nano banana can both do that with images, there’s zero reason to think we can’t train video models for masking.

This feels a lot like critiquing stable diffusion over hands and text, which the new SOTA models all handle well.

One of the easiest iterations on these models is to add more training cases to the benchmarks. That’s a timeline of months, not comparable to forecasting progress over 20 years like fusion.

Re: How the AI Bubble Will Pop

#264

Post by piva00 »

throwaway-0001 wrote: post #201
Diti wrote: post #190

Earlier quoted context omitted.

LLMs cannot think on their own, they’re glorified autocomplete automatons writing things based on past training. If the “AI figured out something on your mind”, it is extremely likely the “thing on your mind” was present in the training corpus, and survivorship bias made you notice.

Tbh if Claude is smarter than average person, and it is, then 50% of the population is not even a glorified auto complete. Imagine that, all not very bright.

They are... People. Dehumanising people is never a good sign about someone's psyche.

Re: How the AI Bubble Will Pop

#265

Post by HarHarVeryFunny »

amelius wrote: post #136
donatj wrote: post #123

I feel kind of like a Luddite sometimes but I don't understand why EVERYONE is rushing to use AI? I use a couple different agents to help me code, and ChatGPT has largely replaced Google in my everyday use, but I genuinely don't understand the value proposition of every other companies offerings. I really feel like we're in the same "Get it out first, figure out what it is good for later" bubble we had like 7 years a…

> It's like letting a small child drive a car. Bad example, because FSD cars are here.

Yeah, and Tesla cross-country FSD just crashed after 60 miles, and Tesla RoboTaxi had multiple accidents within first few days.

Other companies like Wayno seem to do better, but in general I wouldn't hold up self-driving cars as an example of how great AI is, and in any case calling it all "AI" is obscuring the fact that LLMs and FSD are completely different technologies.

In fact, until last year Tesla FSD wasn't even AI - the driving component was C++ and only the vision system was a neural net (with that being object recognition - convolutional neural net, not a Transformer).

Re: How the AI Bubble Will Pop

#266

Post by dist-epoch »

eek2121 wrote: post #175
donatj wrote: post #123

I feel kind of like a Luddite sometimes but I don't understand why EVERYONE is rushing to use AI? I use a couple different agents to help me code, and ChatGPT has largely replaced Google in my everyday use, but I genuinely don't understand the value proposition of every other companies offerings. I really feel like we're in the same "Get it out first, figure out what it is good for later" bubble we had like 7 years a…

In my eyes, it'd be cheaper for a company to simply purchase laptops with decent hardware specs, and run the LLMs locally. I've had decent results from various models I've run via LMStudio, and bonus points: It costs nothing and doesn't even use all that much CPU/GPU power. Just my opinion as a FORMER senior software dev (disabled now).

Electricity is not free. If you do the math, online LLMs are much cheaper. And this is before considering capabilities/speed.

Re: How the AI Bubble Will Pop

#267

Post by greekrich92 »

ivape wrote: post #149
donatj wrote: post #123

I feel kind of like a Luddite sometimes but I don't understand why EVERYONE is rushing to use AI? I use a couple different agents to help me code, and ChatGPT has largely replaced Google in my everyday use, but I genuinely don't understand the value proposition of every other companies offerings. I really feel like we're in the same "Get it out first, figure out what it is good for later" bubble we had like 7 years a…

AI figured out something on my mind that I didn’t tell it about yesterday (latest Sonnet). My best advice to you is to spend time and allow the AI to blow your mind. Then you’ll get it. Sometimes I sit in wonder at how the fuck it’s able to figure out that much intent without specific instructions. There’s no way anyone programmed it to understand that much. If you’re not blown away by this then I have to assume you…

Sincerely, consider that you may be at risk of an LLM harming your mental health

Re: How the AI Bubble Will Pop

#268

Post by dist-epoch »

emp17344 wrote: post #157
ivape wrote: post #149

Earlier quoted context omitted.

AI figured out something on my mind that I didn’t tell it about yesterday (latest Sonnet). My best advice to you is to spend time and allow the AI to blow your mind. Then you’ll get it. Sometimes I sit in wonder at how the fuck it’s able to figure out that much intent without specific instructions. There’s no way anyone programmed it to understand that much. If you’re not blown away by this then I have to assume you…

I don’t understand what you’re saying. You know the AI is incapable of reading your mind, right? Can you provide more information?

> You know the AI is incapable of reading your mind, right.

Of course they can, just like a psychiatrist can.

Re: How the AI Bubble Will Pop

#269

Post by Gerardo1 »

rafaelmn wrote: post #100
cornholio wrote: post #64

Earlier quoted context omitted.

It's silly to say that the only objective that will vindicate AI investments is AGI. Current batch of deep learning models are fundamentally a technology for labor automation. This is immensely useful in itself, without the need to do AGI. The Sora2 capabilities are absolutely wild (see a great example here of what non-professional users are already able to create with it: https://www.youtube.com/watch?v=HXp8_w3XzgU…

If you calculate the investment into AI and then divide by say 100k that's how many man-years need to replace with AI to be cost effective as labor automation the numbers aren't that promising given the current level of capability.

Don't even need to get too fancy with it. Open AI has publicly committed to ~$500B in spending over the next several years (nevermind even they don't expect to actually bring that much revenue in)

$500B/$100,000 is 5 million, or 167k 30-year careers.

The math is ludicrous, and the people saying it's fine are incomnprehensible to me.

Another comment on a similar post just said, no hyperbole, irony, or joke intended: "Just you switching away from Google is already justifying 1T infrastructure spend."

Re: How the AI Bubble Will Pop

#270

Post by Gazoche »

ttoinou wrote: post #25

Every financial bubble has moments where, looking back, one thinks: How did any sentient person miss the signs? Well maybe a lot of people agree already with what the author is saying : the economics might crash, but the technology is here to stay. So we don't care about the bubble

And, conversely, some don't care about the technology but want to ride the bubble and exit right before it pops.
Post reply on HN