Back to list

AI Business虎嗅

AI Companies' IPOs Begin to Hit the Brakes Collectively

虎嗅1 min read
Read original

AI Summary

Recently, US-listed AI-related companies have been postponing or withdrawing IPOs in succession. While private markets still assign sky-high valuations to AI firms, public-market investors are refusing to accept them at the listing stage. Analysts note parallels with the dot-com bubble era of 2000.

AI Insights

The wave of AI-linked IPO postponements (Oura, Firmus, AirTrunk, Holtec) signals that public markets are repricing AI assets, as private valuations have already captured much of the future growth premium. The parallel with the 2000 dot-com IPO withdrawals suggests that companies with strong revenue and blue-chip backers still face valuation pressure when late-stage private rounds outpace what public investors will pay.

Analysis written by AI from the original article. Check it against the source.

Translated text

Recently, an unusual phenomenon has emerged in the US stock market. Private markets still assign sky-high valuations to AI companies, but when it comes to listing, public-market investors are no longer accepting them. Between mid-September and mid-October 2025, at least four companies linked to the AI industry chain or AI investment narratives announced postponements or withdrawals of their IPOs, spanning smart hardware, AI computing, data centers and nuclear power. On October 9, Firmus, an Australian AI data center company in which Nvidia holds a 7.2% stake, formally withdrew its IPO. The company was backed by Nvidia, Blackstone and Coatue, and its customers include OpenAI and Meta. Smart ring maker Oura also suddenly postponed its IPO on the day its issue price was to be set, with the core reason being a failure to agree on valuation. In private markets, AI asset valuations keep climbing, while public markets have begun to say no to these high valuations. A similar pattern appeared in the US IPO market after the Nasdaq peaked in March 2000, when a wave of withdrawals followed within a month. Today's AI assets have not yet reached that stage, but a divergence between private-market valuation expectations and public-market pricing is already visible for some AI assets.

Related News

3