The bubble in AI is not the technology. It is the money being bet on owning the infrastructure underneath it, and a growing amount of evidence that the bet does not pay back the way it is being priced.
Is AI actually a bubble?
The infrastructure spend, yes, very likely. Open-source models coming out of China are already closing the gap to within a few percentage points of the most expensive frontier systems, at a fraction of the cost. When you can get 95 to 98 percent of the result for free, the case for the enormous capital being poured into owning the remaining few percent gets very hard to defend, unless something genuinely new arrives to justify it.
Hasn’t this happened before?
Yes, and the comparison is useful, but not in the way people usually mean it. After the dot-com crash, the companies disappeared, but the physical thing they built, the actual internet backbone, stayed standing and became the foundation everything since has run on. That is not what survives an AI bubble bursting.
So what’s actually left standing, if it isn’t infrastructure?
Belief, and trust. Not a fiber network you can point to, something far harder to build and far easier to lose: the conviction, at both a personal and a business level, that this genuinely changes what is possible, new revenue, new markets, new ways of working, and that it is worth staking real decisions on. That is a strange thing for a bubble to leave behind, but it is also, arguably, the more durable one. Trust does not depreciate the way hardware does.
What does that actually mean for how a business should be reacting to all this right now?
The businesses that move fastest through this period, using what is genuinely working rather than waiting for certainty, get a real head start on everyone still watching from the sidelines. And the size advantage that used to matter is inverting: a one or two person company, moving fast with the right tools, is a real threat to organizations many times its size in a way that was not true even two years ago.
Is AI a bubble? Possibly the infrastructure is. What survives it is not going to look like what survived the last one, and that is worth thinking about now, not after it pops.
If you are leading your organization through this, this is the kind of work I do with a limited number of senior leaders each quarter. → https://anglero.com/the-reset/
Strategic AI Advisor, MerkabaPhi AS, Oslo. 450+ keynotes across 30+ countries.
Listen to the Clarity at the Top podcast on Spotify.
Inquiries: anglero.com