Alphabet's cash burn raises alarm for Big Tech as AI spending climbs • Hacker News Board

Re: Alphabet's cash burn raises alarm for Big Tech as AI spending climbs

#261

Post by fooker » Thu, Jul 23, 2026, 6:32 PM UTC

maxglute wrote: post #239

fooker wrote: post #158

Earlier quoted context omitted.

H100 is nearing five years and costs more to buy a used one now than a new one when it was released :) You are completely missing the bet these companies are making. They think can outlast their competitors and capture a larger portion of the pie while the cost of inference keeps going down dramatically. If you haven't been paying attention, the cost is about 1/100th of what it was in 2024. This is the trajectory pre…

The fact that supply constrained GPUs holding value is negative indicator. It's like the tulips mania except bubble enthusiasts are buying wilted/dead tulips because live tulips supply constrained due to irrational demand. Now GPUs has more gross utility than tulips but seems like at current revenue/capex spend, every GPU is still negative net financial yield - they lose money - literally buying tulips and watching i…

You're making a classic philosophical error here by accidentally anthropomorphizing companies :)

A company losing out on a risky bet and failing, letting a new upstart rise up is pretty natural. A large fraction of experts from the failed companies continue at the new ones, business as usual. There are engineering teams at $BIGTECH now full of OS, database, or compiler experts from XP, Sun, HP etc.

This natural ability of companies to take risky bets is what made silicon valley successful.

Re: Alphabet's cash burn raises alarm for Big Tech as AI spending climbs

#262

Post by lardosaurusrex » Thu, Jul 23, 2026, 6:36 PM UTC

gmadsen wrote: post #260

lardosaurusrex wrote: post #209

Earlier quoted context omitted.

I think this fails to take into account how many people are fine with "fast enough" vs "fastest". I've seen people happily use AI that takes several minutes to generate text or edit an image because to them they already aren't using their computer when they tell it to start; they just grab their phone and walk away and come back only to check in on it. I feel like people here and on other technology discussions -- al…

That is completely discounting future capabilities and new use cases. Sure in 10 years you will have current SOTA locally, but in no way is it obvious we are anywhere near the limit of marginal value from improved capability

Unless you're a developer or doing complicated research you do not need much to use an LLM at home with consumer-grade hardware.

I'm editing photos with 32GB of DDR4 RAM at 3200mhz alongside a 3050 with a nearly decade-old mobo and ryzen 5800XT and at most? It takes 30 seconds and that's allowing the gpu to use my actual RAM as 'fallback' memory.

There are photoshop filters -now- that take longer than that before the LLM craze even began. Hell; I can train a lora in an hour or two.

This obsession with "more specs more data faster and faster" isn't going to win; it already isn't winning.

Deal with it or get wrecked.

Re: Alphabet's cash burn raises alarm for Big Tech as AI spending climbs

#263

Post by CuriousSkeptic » Thu, Jul 23, 2026, 6:54 PM UTC

mcbuilder wrote: post #140

LunaSea wrote: post #119

Earlier quoted context omitted.

Why? GPUs are replaced every 3 to 5 years. This is going to be an ongoing operational cost forever. It will probably increase more if larger models require bigger VRAM sizes.

We have probably hit a limit to scaling LLMs through raw parameter count alone, at least we're not seeing the exponential pace. I personally think we'll end up with a nice sigmoid curve plateauing in the sub 10T parameter regime. The amount of tokens processed (in inference) is increasing exponentially though (I've been following open router usage stats for years and it's always been exponential). We will of course m…

> We have probably hit a limit to scaling LLMs through raw parameter count alone, at least we're not seeing the exponential pace.

Source?

Re: Alphabet's cash burn raises alarm for Big Tech as AI spending climbs

#264

Post by hnfong » Thu, Jul 23, 2026, 6:59 PM UTC

mywittyname wrote: post #154

johndough wrote: post #149

Earlier quoted context omitted.

> GPUs become obsolete in 5 years The GPUs are far from worthless after 5 years. E.g. the A100 80GB PCIe version cost around $15k when it was introduced in 2021 and now sells for $10k used. Things might be slightly worse for the data center servers, but I am sure they will find find buyers.

How much of that is due to inflated RAM prices though? I wouldn't assume the current trend is going to continue.

It really depends on how you view the future demand for compute, which really is the crux of the question of whether the capex is rational or not...

Re: Alphabet's cash burn raises alarm for Big Tech as AI spending climbs

#265

Post by hnfong » Thu, Jul 23, 2026, 7:04 PM UTC

hluska wrote: post #222

Imustaskforhelp wrote: post #193

Earlier quoted context omitted.

For what its worth, SK Hynix CEO has every incentive to show that RAM prices will remain high for as long as possible because that is the only thing which is floating their evaluation to such astronomical amounts. Independent estimates sort of show around 2027-28 from what I remember. I remember reading some article which said that RAM prices are already going down from its peak slowly (IIRC I can be wrong, I usually…

[flagged]

It's not directly relevant, but the company has been involved in multiple criminal investigations with alleged price fixing etc.

https://en.wikipedia.org/wiki/DRAM_price_fixing_scandal

Also, I don't recall any era where "proof" is required for allegations. "Evidence" is probably prudent, but proof is usually produced when the accused gets ends up in court.

Unless you time travelled from a feudal era where they throw you in jail for slandering the nobleman or something.

Re: Alphabet's cash burn raises alarm for Big Tech as AI spending climbs

#266

Post by Imustaskforhelp » Thu, Jul 23, 2026, 7:32 PM UTC

justsid wrote: post #251

fooker wrote: post #158

Earlier quoted context omitted.

H100 is nearing five years and costs more to buy a used one now than a new one when it was released :) You are completely missing the bet these companies are making. They think can outlast their competitors and capture a larger portion of the pie while the cost of inference keeps going down dramatically. If you haven't been paying attention, the cost is about 1/100th of what it was in 2024. This is the trajectory pre…

Isn’t that the same bet that famously profitable companies like Uber did in the ride share market?

Uber is a two sided marketplace and was still famously unprofitable for at the same time.

Over 14 years, Uber burnt ~31 Billion dollars and in nearly whereas the amount invested within AI seems to be within Trillions at this point and in near future with a product which doesn't have much moat and shaky financials on profit.

Re: Alphabet's cash burn raises alarm for Big Tech as AI spending climbs

#267

Post by piguin » Thu, Jul 23, 2026, 7:35 PM UTC

itkovian_ wrote: post #250

piguin wrote: post #226

Earlier quoted context omitted.

It will obviously be a large part of the economy like online shopping is today. The companies that built up a lot of debt to be brand names in online shopping primarily went bankrupt because new companies had no debt (and perhaps no negative sentiment from early customer experiences.)

the difference is all the current capex is going to durable, hard to get physical assets + things like PPAs. In your online shopping analogy, the hyperscalars are acting like Amazon in 1998

What destroyed hardware manufacturers after the first bubble burst was all the old but overpowered server hardware ending up in resale after bankruptcies. Hard to get hardware is a bad investment that is also potentially obsolete after new optimizations make the next generation much more efficient. The companies that think their individual optimizations are going to outpace industry wide optimization are delusional, historically speaking.

Re: Alphabet's cash burn raises alarm for Big Tech as AI spending climbs

#268

Post by Imustaskforhelp » Thu, Jul 23, 2026, 7:38 PM UTC

itkovian_ wrote: post #250

piguin wrote: post #226

Earlier quoted context omitted.

It will obviously be a large part of the economy like online shopping is today. The companies that built up a lot of debt to be brand names in online shopping primarily went bankrupt because new companies had no debt (and perhaps no negative sentiment from early customer experiences.)

the difference is all the current capex is going to durable, hard to get physical assets + things like PPAs. In your online shopping analogy, the hyperscalars are acting like Amazon in 1998

> the hyperscalars are acting like Amazon in 1998

I hope you remember furniture.com, pets.com, webvan.com, kozmo.com and many others.

Amazon.com (during 1998) in some sense is the exception, not the norm from the bloodbath in stock markets during the dot com bubble.

(I highly recommend the book How the internet happened for more knowledge about things before, during and after the dot com bubble.)

Re: Alphabet's cash burn raises alarm for Big Tech as AI spending climbs

#269

Post by Imustaskforhelp » Thu, Jul 23, 2026, 7:52 PM UTC

hluska wrote: post #222

Imustaskforhelp wrote: post #193

Earlier quoted context omitted.

For what its worth, SK Hynix CEO has every incentive to show that RAM prices will remain high for as long as possible because that is the only thing which is floating their evaluation to such astronomical amounts. Independent estimates sort of show around 2027-28 from what I remember. I remember reading some article which said that RAM prices are already going down from its peak slowly (IIRC I can be wrong, I usually…

[flagged]

> Back in the day, we used to have this thing called “proof” before making wild allegations of illegality

I understand that you might've wanted to know more sources but its better to ask them directly in good faith as I am more than happy to provide sources and have a good faith discussion! :)

> are you just playing pretend?

(This part feels a bit personally rude to me. I would prefer if you could omit this part of the messages in future or take care of that, Hackernews is a lovely community for the most part so lets keep it that way :-D)

I appreciated reading the comment of @hnfong. The DRAM Price fixing scandal and multiple criminal investigations are something that I didn't even know of!

The general consensus is that its late 27-28 and I would like to share some of the sources that you were asking.

I don't know if they are lying or not but the interview on reuters was conducted on the same day that they went on Nasdaq

> "We forecast that next year will be the worst year in the industry's history from the supply perspective," Kwak told Reuters in an interview on Friday, the day the South Korean memory chipmaker began trading on Nasdaq.[From (1)]

> Multiple analysts push meaningful relief to late 2027 at earliest — with some estimates stretching into 2028[0]

> UBS likewise expects the global DRAM industry to remain undersupplied until at least the second quarter of 2028.[1]

The above one [1] reference is literally from the reuter article itself which was referenced in the parent comment

> Across the U.S., China and Europe, DDR5 RAM prices fell sharply. For example, after reaching an all-time high of $500, Corsair’s Vengeance 32GB kit is now down to $369 on Amazon. Memory has fallen more than 20% from recent peak prices, and stock prices of major memory factories like Micron, Samsung and SK Hynix have seen drops of over 20%.[2]

[0]: https://iqondigital.com/learn/pc-news/when-will-ram-prices-d...

[1]: https://www.reuters.com/world/asia-pacific/sk-hynix-ceo-sees...

[2]: https://www.tomsguide.com/computing/hardware/ram-prices-are-...

have a nice day with the sources and let me know if you want a good faith discussion!

Re: Alphabet's cash burn raises alarm for Big Tech as AI spending climbs

#270

Post by lenerdenator » Thu, Jul 23, 2026, 7:57 PM UTC

zdragnar wrote: post #243

lenerdenator wrote: post #240

Earlier quoted context omitted.

> How long will a SOTA model be necessary? If day to day work can be achieved on an open weight model, the most evaporates overnight. Depends on the advantage it gives people and marketing of that advantage. You'd be surprised at how overpowered the average workplace laptop is. Each company I've been at has had at least some people who do non-technical roles using high-end hardware. Why? Because the account executive…

Execs getting what they want doesn't mean they let everyone have the same thing. It's far more likely that everyone else is using lesser equipment. We have already seen tech workers at big name companies get whiplash from "leaderboards showing people using the most tokens!" as a good thing one month to being pressured to using fewer tokens a month later.

> Execs getting what they want doesn't mean they let everyone have the same thing. It's far more likely that everyone else is using lesser equipment.

You'd be surprised. I've seen people request upgrades and get them. Not even execs, just employees. Even the average devices are probably overpowered these days. Chromebooks could do the vast majority of work in a corporate environment. Try getting workers to accept them, though.

Regardless, there are probably enough reasons to use proprietary Western AI for the time being, particularly in regulated industries, that poop won't completely hit the fan. Particularly if the CTOs start searching the phrase "Operation Aurora".

> We have already seen tech workers at big name companies get whiplash from "leaderboards showing people using the most tokens!" as a good thing one month to being pressured to using fewer tokens a month later.

Indeed. What do you do when you want a group of people to do something that they'd be otherwise adverse to doing at all? You rank them on it and let them get in a competition over who can do it the most. That's what tokenmaxxing was about: getting people to use the AI at all. Now that the people are using AI, we move onto another objective, which is getting them to use the tokens efficiently. The trick is measuring that efficiency.