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

AI's debt binge can't last, hidden borrowing reaches $1.65T

fortune.com

Post reply on HN
Print view View original

131–140 of 189 posts

Re: AI's debt binge can't last, hidden borrowing reaches $1.65T

#131

Post by seizethecheese »

billywhizz wrote: post #80
seizethecheese wrote: post #8

> AI’s insatiable need for debt has so far been matched by investors’ appetite for it, but they may turn nauseous on the belly-busting volumes coming from tech giants. Headline doesn't really match the facts in the article. The article seems to say "hyperscalers are borrowing an enormous amount and so far people are lending to them. Other people are worried that this will stop".

"people are lending to them" is doing a lot of heavy lifting here. e.g. https://prospect.org/2026/08/03/ai-bailout-could-be-baked-in...

The article transparently documents that people are lending to them.

Re: AI's debt binge can't last, hidden borrowing reaches $1.65T

#132

Post by seizethecheese »

kube-system wrote: post #16
seizethecheese wrote: post #8

> AI’s insatiable need for debt has so far been matched by investors’ appetite for it, but they may turn nauseous on the belly-busting volumes coming from tech giants. Headline doesn't really match the facts in the article. The article seems to say "hyperscalers are borrowing an enormous amount and so far people are lending to them. Other people are worried that this will stop".

> "hyperscalers are borrowing an enormous amount and so far people are lending to them. Other people are worried that this will stop" You say this as if when "lending stops", it isn't a big deal. What you're describing is a concern for a collapse in finance markets.

Sure, I’m just saying the article doesn’t have any evidence of the lending stopping.

Re: AI's debt binge can't last, hidden borrowing reaches $1.65T

#133

Post by seizethecheese »

bigbuppo wrote: post #14
seizethecheese wrote: post #8

> AI’s insatiable need for debt has so far been matched by investors’ appetite for it, but they may turn nauseous on the belly-busting volumes coming from tech giants. Headline doesn't really match the facts in the article. The article seems to say "hyperscalers are borrowing an enormous amount and so far people are lending to them. Other people are worried that this will stop".

Historically speaking, when the lending spigot is turned off it happens suddenly. But hey, it will be different this time. My future's so bright I gotta' wear million dollar shades.

It clearly will be a big problem when the lending stops, there’s just no evidence of it in the article.

Re: AI's debt binge can't last, hidden borrowing reaches $1.65T

#134

Post by jdub »

minimaltom wrote: post #55
metalliqaz wrote: post #43

Leverage is imploding the Korean market, and so it will be for the US.

South Korea had an extremely large population of retail investors investing in options and leveraged ETFs, to the point that 3% of the adult population has now been margin-called. That setup isnt true for the US, not even close.

Are you sure? What is the 401(k) exposure to "AI" related stocks and broader market shocks?

Re: AI's debt binge can't last, hidden borrowing reaches $1.65T

#135

Post by bigstrat2003 »

afry1 wrote: post #73
mapping365 wrote: post #70

The people who made money on fiber and railroads were the inheritors after the timeline mismatch bankrupted the original players who did the investment. Even if AI turns out to be everything it promises, you can mistime the investment and lose everything.

Fiber and railroads don't depreciate after 3 years of use like AI chips. Fiber and railroads don't need tens of billions of dollars in continuing yearly maintenance expenses to keep them from going stale.

Fiber and railroads also actually accomplish useful things and aren't just a hype vehicle.

Re: AI's debt binge can't last, hidden borrowing reaches $1.65T

#136

Post by seizethecheese »

keeda wrote: post #130

Genuine question: these companies had double-digit billions of free cash flow per quarter, about $0.3T a year aggregate, before the AI boom started and they began splurging on CapEx; is the $1.65T number that bad in that context? Let's assume the extreme worst case scenario where the bubble pops so comprehensively that the entire AI business is written off, without any change to the debt owed, and these companies ret…

I think the stock wipeout would itself trigger a recession.

If the hyperscalers needed to wipe out most of their income on interest expense they’d lose a large amount of their market capitalization. This could drop the stock market a huge amount, and a lot of spending is driven by the “wealth effect” of households feeling wealthy.

Re: AI's debt binge can't last, hidden borrowing reaches $1.65T

#137

Post by saulpw »

dom96 wrote: post #85
echelon wrote: post #38

Earlier quoted context omitted.

The internet remains the biggest singular development of my entire life. The most valuable companies in the world are internet companies. Journalists have been eager to call AI "over" since 2022, and yet: - Models just got good at writing code this year - Models just got good at editing images last year - Models just got good at cinematic video this year This hasn't even played out. It hasn't even started. Why on ear…

> Models just got good at writing code this year That's not correct, is it? Opus 4.5 came out in Nov 2025. Some might say models were good at coding even before that.

That's "this year" [the past 12 months].

Re: AI's debt binge can't last, hidden borrowing reaches $1.65T

#138

Post by seizethecheese »

afry1 wrote: post #73
mapping365 wrote: post #70

The people who made money on fiber and railroads were the inheritors after the timeline mismatch bankrupted the original players who did the investment. Even if AI turns out to be everything it promises, you can mistime the investment and lose everything.

Fiber and railroads don't depreciate after 3 years of use like AI chips. Fiber and railroads don't need tens of billions of dollars in continuing yearly maintenance expenses to keep them from going stale.

The shorter useful life implies less risk though, since therefore the projections don’t have to play out for very long.

Re: AI's debt binge can't last, hidden borrowing reaches $1.65T

#139

Post by afry1 »

KennyBlanken wrote: post #127
afry1 wrote: post #73

Earlier quoted context omitted.

Fiber and railroads don't depreciate after 3 years of use like AI chips. Fiber and railroads don't need tens of billions of dollars in continuing yearly maintenance expenses to keep them from going stale.

> Fiber and railroads don't need tens of billions of dollars in continuing yearly maintenance expenses to keep them from going stale. Congrats, you're as qualified as all the private equity companies that have been piling into railroads! That's precisely the sort of attitude that got us East Palestine (and others.) BNSF (for example) spends billions of dollars a year on maintenance. When trains go over rails enough i…

Infrastructure depreciates!

More slowly and with less expense in relative terms than an LLM model.

You get a lot more bang from your buck from a 10 year old set of under-maintained railroad tracks than you do from a 10 year old unmaintained LLM.

To compare the depreciation or ongoing maintenance expense of the two as apples to apples is ludicrous. To equate the capital expenditure and long term value of the two as equal is also ridiculous.

Re: AI's debt binge can't last, hidden borrowing reaches $1.65T

#140

Post by kube-system »

seizethecheese wrote: post #132
kube-system wrote: post #16

Earlier quoted context omitted.

> "hyperscalers are borrowing an enormous amount and so far people are lending to them. Other people are worried that this will stop" You say this as if when "lending stops", it isn't a big deal. What you're describing is a concern for a collapse in finance markets.

Sure, I’m just saying the article doesn’t have any evidence of the lending stopping.

By the time we have that evidence, the crash is already beginning. What we have today is evidence that lending is overextended unless AI hyperscalers start to make huge profits.
Post reply on HN