TL;DR
OpenAI and Anthropic rely on venture-funded AI startups for 80% of enterprise revenue, and their trillion-dollar compute commitments resemble subprime ARMs, creating huge concentration risk.
AI leaders shout that AGI has arrived, but the truth is that the two biggest AI companies are sitting on a volcano that could erupt at any moment.
Revenue kept alive by transfusion
OpenAI and Anthropic get 80% of their enterprise revenue from 1% of their customers, and most of those are venture-backed AI startups. These startups burn investors' cash, subsidize users' insane token usage, and pass the money straight to the AI labs. If funding dries up, so does that revenue.
That's like the NINJA loans before the financial crisis — no income, no job, but still got a mortgage. AI startups are today's NINJA borrowers.
Compute promises are a ticking bomb
Worse still, the two labs have signed take-or-pay compute deals worth over $1.3 trillion, with more than $200 billion falling due each year starting in 2027. Yet they themselves lose tens of billions annually.
It's like the adjustable-rate mortgages of the subprime crisis — teaser rates look great for a few years, then the reset hits and monthly payments explode, and the borrower defaults.
Who is swimming naked?
Nvidia's customers also depend heavily on these two labs for revenue, which in turn depends on them keeping paying. The whole chain is interlinked at every point, so one missed payment and the bubble pops.
Don't listen to the 'don't think too hard, just look at the growth rate' crowd. When money stops being free, it'll all come out in the wash.
Curated from high-quality sources, with concise summaries and key takeaways.
