Countdown until the AI bubble bursts | Hacker News Reader

Countdown until the AI bubble bursts

pop-the-bubble.xyz

78 pointstapematch36 comments

I think you should have all the different models complete the same task and get different clocks going for each. That might be funnier.

It'd be kinda funny if they asked each AI every day and updated the clocks/rationale accordingly.

They do (with only one AI though): "How does it work? Every night, a script runs that asks Gemini to search the web for the latest AI news—both hype and criticism. Based on the sentiment and economic indicators it finds, it updates the predicted "burst date" and explains its reasoning."

the LLM doesn't have the concept of time and it doesn't incorporate new data, unless it's put into the context, so I don't see the point of this suggestion.

The whole thing is a joke anyways. The numbers are meaningless. The inconsistency in output from day to day highlights that even more.

Please add a graph of the prediction history.

> The Predicted Date: October 3, 2026

> — Analysis generated by Gemini 2.5 Flash on December 4, 2025 ...

Also:

> Registry Expiry Date: 2026-10-01T23:59:59.0Z

This domain will be the pop heard around the world

Well the date is going to change everyday based on what Gemini says

I see, there's hope yet. I didn't read that close, gotta skim with all this ~value~ generation

I mean, I got the gist it might change... but I assumed this is a flat page, made once. For the lulz. Entirely missed the relevant FAQ!

edit: I'll actually check this again to see what the robot(s) think, others have good a suggestion: more, compare sentiment beyond Gemini.

Yeah it would be better to have that it is AI generated at the top of the page. I also think comparisons would be hilarious

the date has been changing for years now. one day someone will actually define what bubble we are waiting to burst and then will poke it to burst and many will say “hey, see I was right all along” ;)

And now we’ll be able to say Gemini was right

Of course … :)

Would be awesome to have logs of the daily estimates and their reasons

Ok I have been watching this and came back to it. The predicted date keeps getting moved up so much so it is now predicted for Dec 17, 2025! olol

As of today ChatGPT is saying September 15, 2026 +/- 3 months

Well this is quite the cute little paradox.

Bubble pops? The AI was right! AI works! Doubters were wrong, everybody should be throwing their money back in! That wasn't a bubble that was Future Earnings.

Bubble doesn't pop? AI fails and it's a scam & a bubble! Why are we all still pouring money into this bubble? Someone in the comments, in response, moves the goalposts on AI.

Heavy dose of /s, of course.

It just needs to pop at a different time than predicted.

Why sarcasm? It is an interesting paradox.

Is there a place to donate for the cause?

NYSE. Pump those market caps up.

I guess I wonder what people think the AI bubble is.

Are you worried about the high valuations of the big AI companies (OpenAI, Nvidia, etc)? Then sure, that will correct over time. There will likely be 1-2 big winners.

But if you're talking about AI in general... right now is the least amount of money companies will be spending on AI ever. It will only go up. This isn't crypto.

> This isn't crypto.

Bitcoin is at $92k.

Sure, but crypto companies aren’t as in vogue anymore.

AI companies will someday just be called companies, much like how tech companies are just companies.

...do you not see the parallel in the words you just wrote? That AI's value will continue increasing even as the buzzwords fall out of the public consciousness, just as crypto continued to gain value even as the buzzwords fell out of the public consciousness?

(Paraphrase != endorsement.)

I think when most people talk about the "AI bubble bursting", they mean a dramatic end to this notion that AI is the "next big thing". Much like how we had Web3 and NFTs and all those other things that were going to change how we interacted with the internet.

Sure, my BTC is up, but I can go weeks or months without interacting with a blockchain in any way, directly or indirectly.

Valuations of individual AI companies might (and will) drop, but we are currently experiencing the least amount of AI in our everyday lives that we (or our children) ever will.

Crypto has had a strong boom bust cycle, dropping ~80%+ every four years or so.

But if the value add of AI fails to materialize like the current valuations are based on (as in the % of GDP impacted), then you’re going to see investors losing money on the investments.

I wonder if the max pain AI bubble burst is just the mag7 being tightly rangebound for years, down in real terms but buoyed by the global money printers switching on again to repair the clearly distressed global economy. This outcome would disappoint the AI haters and the true believers alike.

you can write all you want about an AI bubble but when I can do from the command line with agentic AI changes to a large code base that would take over a year and get it done in a week or two, it's simply a dramatic evolution of software engineering by one competent software engineer

> you can write all you want about an AI bubble

Bubble =/= useless

Bubble means over-hyped

How can the predictive market bros quantifiably turn this into a bet? At what point can we verifiably say we heard a loud pop?

When investors start pulling Waring Hudsuckers.

> How can the predictive market bros quantifiably turn this into a bet?

Put options trading.[^1]

> At what point can we verifiably say we heard a loud pop?

Definitely after the fact. Maybe months, maybe years. I have personally been following the parallels with the dot-com bubble burst, but it's impossible to time the market. Financial data is always too slow when you want it fast and too fast when you need to watch it carefully.

[^1] https://www.investopedia.com/ask/answers/06/sellingoptions.a...

> At what point can we verifiably say we heard a loud pop?

Usually, well after the fact. Especially in a case like this, where most of the money is being spent by the hyperscalers, who are unlikely to outright _collapse_ or anything. There are exceptions, where a bubble popping is accompanied by a massive single-day market crash, but that's not _really_ the norm.

Polymarket:

This market will resolve to "Yes" if the AI industry experiences an industry downturn by the specified date, 11:59 PM ET. Otherwise, this market will resolve to "No".

For the purposes of this market, the AI industry will be considered to have experienced an industry downturn once at least three of the following events have occurred within 90 days of this market's specified timeframe: - NVIDIA Corporation (NVDA) closing stock price is down 50% from its all-time high. - iShares PHLX Semiconductor ETF (SOXX) closing stock price is down 40% from its all-time high. - OpenAI, Inc. or Anthropic PBC declares bankruptcy. - OpenAI, Inc. is acquired. - H100 rental price falls to $1.00 or lower for five consecutive days, as shown on the SiliconData Silicon Index at: https://www.silicondata.com/products/silicon-index. - Major AI Hardware Supplier Collapse: Taiwan Semiconductor Manufacturing Company Limited (TSM), ASML Holding N.V. (ASML), Broadcom Inc. (AVGO), Arista Networks, Inc. (ANET), or Super Micro Computer, Inc. (SMCI), closing stock price is down 50% from its all-time high.

Current odds - 35% by end 2026.

Reply on news.ycombinator.com