[JumpCrisscross]

> traditionally the pop is delayed while those who realized most of the gains attempt to offload the risk to other parties

What? Source? Plenty of investment bubbles pop before the bag is passed.

This thread involves a lot of people looking at something they don't like and presuming karmic forces will give them what they deserve. There is no reason these companies, even if massively overvalued, have to "pop."

That's fundamentally different from e.g. the financial crisis, or the 2023 bank collapses, or even the dot-com bubble. Those did not have the ability to self correct. There was no slow deflation other than through a bailout.

[overfeed]

> There is no reason these companies, even if massively overvalued, have to "pop."

This is a wild thing say without any qualification.

[JumpCrisscross]

> This is a wild thing say without any qualification

It’s really not. Bubbles are notable because most elevated asset prices slowly go down. And they have common characteristics that force the reckoning. Usually debt. Sometimes operational leverage.

[marcus_holmes]

I'm genuinely curious why you say this is different from the dot-com bubble?

As I see it, this is the exact same situation - wildly overvalued companies based on investor exuberance, the underlying business is not capable of supporting this kind of valuation. IPO tends to be the crunch point at which this overvaluation is exposed. Once exposed, the valuation correction spreads to other similar businesses quickly and the bubble pops.

What's the self-correction ability that AI companies have?

[JumpCrisscross]

> genuinely curious why you say this is different from the dot-com bubble?

A lot more revenue. Dot coms were going public pre revenue. And Anthropic is profitable. Both it and SpaceX wouldn’t be dependent on further stock sales to stay alive—that lets them weather a downturn.

[marcus_holmes]

As I understand the situation, Anthropic is revenue-positive but not profitable. As usual, Ed Zitron covers this well [0].

As with the dot-com bubble, there is a lot of voodoo accountancy (and flat-out lies) about the actual situation here.

As I understand it, the basic problem is that the big three can't charge enough per token to cover costs because they're in competition with each other (and one of those is Google that can afford to buy market share using its other operating revenues), and the OSS/cheap Chinese models.

And this situation is unlikely to get better in the short term because building cheaper per-token capacity is very expensive and time-consuming.

[0] https://www.wheresyoured.at/anthropics-profitability-swindle...

[JumpCrisscross]

> this situation is unlikely to get better in the short term because building cheaper per-token capacity is very expensive and time-consuming

They don’t need to fix it in the short term.

Look, this could be total nonsense. But what won’t happen is Anthropic or SpaceX disappearing inside a year. That was true in the 90s because the only cash flow going into those companies came from investors.

[marcus_holmes]

I notice you left out OpenAI from that ;)

Agree, some of these are valid businesses. But they are also massively overvalued on that underlying valid business, because of investor enthusiasm. When the bubble pops they are going to have real problems because of that overvaluation. Hopefully they survive, as a lot of the dotcom businesses did.

I think the real bloodbath will be the second-tier businesses that are mostly reselling cheap tokens to a market niche with custom prompts, and also massively overvalued as "AI businesses". And that kinda mirrors what happened in the dotcom bust - all the overvalued "webscale" businesses that hadn't really worked out a solid model yet went to the wall immediately

[JumpCrisscross]

> notice you left out OpenAI from that

OpenAI seems to have made debt-like commitments to spending on infrastructure. If those are indeed binding, they may have less flexibility than the others. (If Anthropic’s revenue growth stalls and its valuation halves, it should still be a going concern.)

[TaupeRanger]

If what you say is true and it predicts the future, then everyone would be selling right now. The fact is, no one knows when or if the bubble will pop, and we will only be able to say in hindsight whether your comparison is correct.

[marcus_holmes]

There will come a point, probably in the next year or two, when investors realise this is true and it predicts the future. And then they will start selling. And that's how the bubble pops.

Bubbles always pop. The question is only ever about the timing.

[michaelmrose]

I said attempt to offload see mortgage backed securities for one such attempt.

The point is that nobody wants to be the first out of a hot market nor the last so that bubbles everyone knows are bubbles first hang on despite it being broadly believed to be so and then crash as people head for the exits.

Broadly people are taking on debt to realize profits that may not exist. Retrospectively widely acknowledged bubbles like every crash in the last century all popped im not aware of any big enough to cause a recession that petered out slowly. Since we don't need to look up 100 years of crashes together can you name some similarly large issues that were resolved slowly over time?