Markets JUN 29, 2026

OpenAI Pushes IPO to 2027 as Altman Holds Out for $1T Valuation

Sam Altman is leaning past a 2026 debut after SpaceX's post-IPO slide spooked advisers; SoftBank fell the most since August 2024 and the Nasdaq sank a sixth straight session.

BY HENLEY MARRAST · MARKETS DESK

OpenAI is leaning toward delaying its public offering until 2027 rather than accept a 2026 debut priced below $1 trillion, according to a New York Times report citing three people involved in the deliberations. The company confidentially filed its S-1 with the SEC on June 8. Sam Altman, by this telling, would rather wait out the tape than print a number that doesn’t begin with a “T.”

The timing is unsubtle. SpaceX’s June 12 IPO debuted “blistering” and then unwound, with shares dipping below the $150 listing price on Tuesday and more than $915 billion in value erased from the peak. The Nasdaq is set to fall 1.2% Friday, its sixth straight lower session, and it’s now off more than 6% from its June 2 high. Microsoft and Meta have slipped into bear-market territory; Amazon, Apple, Alphabet, Nvidia, and Tesla are in correction. On Tuesday, Sandisk, Micron, and Arm fell more than 10% in a session; Nvidia dropped 4.15%.

This is the window into which OpenAI was supposed to price.

The collateral damage runs through Tokyo. SoftBank, which co-led OpenAI’s $122 billion March round and has put roughly $65 billion into the company for a 13% stake (second only to Microsoft’s 27%), saw its shares fall the most since August 2024 on the delay report. A margin loan of at least $6 billion backed by SoftBank’s OpenAI equity is reportedly stalled in negotiations. The IPO isn’t just an exit; it’s the mark that unlocks everyone else’s balance sheet.

Prediction markets are pricing the patience. Kalshi puts the odds of an IPO announcement by March 1, 2027 at 59%, by June 2027 at 73%, and at roughly one-in-three before January 1. JPMorgan analysts have weighed in on the same arithmetic.

Altman’s bet is that holding out preserves the number that defines the category. The risk is that the tape decides what a frontier model is worth before he does.

Sources

About the author

Henley Marrast

MARKETS DESK

Henley Marrast covers AI-equity flow, accelerator demand, and earnings prints for AI Sheet Report. She leads coverage of the public AI complex from the New York markets desk, with a focus on the daily tape and quarterly results. She has been writing about technology markets for several years.