AI IPO Fail for NVIDIA Backed Firmus - OpenAI is DEAD

A 21-minute monologue, uploaded on 9 October 2026, the day Firmus pulled its IPO. Eli reads the news (Firmus is the Nvidia-backed Australian AI data-centre company) and asks one question: if the story is true, why isn’t this the best moment in history to go public? We’ve been told AI will print money, cure disease and solve science, and that compute is as good as gold, to the point that people are trying to make GPUs tradeable. Yet Firmus walked away from what would have been one of the largest share sales in Australian history: about US$5bn raised at A$11 a share, valuing the company at about US$30.6bn (A$43.7bn). Its board said the terms didn’t reflect the business, and it will raise privately instead. Reporting the same day put it down to weak demand.

He calls himself the man in the middle, which annoys both sides. The technology is fine. It just isn’t worth trillions, any more than SQLite, the most widely deployed database in the world, would be. He lists the cracks: Oura shelved its IPO at the end of September, other listings were pulled or delayed, OpenAI’s float is now expected in 2027 while it raises about $30bn privately, and the “tokenmaxxing” fashion (leaderboards for burning tokens) has given way to companies capping AI spending. Anthropic is aiming for about $2 trillion at its IPO, now expected in November. Eli credits Michael Burry with a thesis that it has to list before its numbers turn; the most the fact-check found from Burry is a hope that a market drop would stop both the OpenAI and Anthropic IPOs.

His worry is the private market. Each funding round should have a new lead investor, so fresh eyes test the price. WeWork reached $47bn because SoftBank led round after round. If every private investor is “AI-pilled”, new eyes add nothing. He gives two more examples. OpenAI’s release of manuscripts claiming more than 300 open maths problems makes him ask what those are worth in money: maths doodles rather than fusion. And SpaceX’s valuation is mostly the aura of Musk, which neither Dario Amodei nor Sam Altman has, so pure AI companies may list badly. The title’s “OpenAI is DEAD” isn’t argued in the episode.