OpenAI is in early-stage talks to raise private capital at a valuation of about $1.2 trillion, with the company reportedly seeking as much as $1.5 trillion — figures that would make it one of the most valuable enterprises ever assembled, public or private.
The framing matters as much as the number. In most large private rounds, the company hires bankers and courts institutional money. In this case, the conversations were initiated by investors rather than by OpenAI, an inversion that says as much about the scarcity of large-scale AI assets as it does about OpenAI itself. The figures are also potential valuations under discussion, not capital raised, and remain subject to diligence.
The step-up is steep. OpenAI was valued at $500 billion in October 2025, then at $852 billion in March 2026, when it raised $122 billion in a combination of equity and debt led by Nvidia, Amazon and SoftBank. A $1.2 trillion valuation would represent a 41% increase from March; $1.5 trillion would be a 76% increase.
The backdrop is a company deliberately choosing private markets over public ones. In an interview with Fortune, Altman said "right now would be an ill-advised moment to go public," citing risks that AI systems could make unchecked decisions causing measurable harm. Asked whether a 10% probability of catastrophic outcomes was a fair working figure, he declined to confirm a specific number, but said: "Whether it's 10 or eight or six, the point is ... we need to act such that we are not taking any of those numbers of risk." Altman has ruled out an IPO this year and named no replacement date, though OpenAI's chief financial officer has previously pointed employees toward 2027 or sooner. Rival Anthropic's listing is still expected this year.
Analysts read the round as a capital raise rather than price discovery. "Altman ruled out a 2026 IPO days before this leaked, and OpenAI spent about $34 billion last year, so read this as a capital raise," said Harrison Rolfes, a senior research analyst at PitchBook. "The $1.2 trillion is what investors offered; the $1.5 trillion is what OpenAI wants, and nothing is priced." He noted that with OpenAI's annualized revenue run-rate reported above $40 billion, the range implies roughly 30 to 38 times revenue — similar to the multiple Anthropic is asking at a reported $2 trillion valuation on a faster-growing business that already reports positive adjusted operating income.
Rolfes also drew a pointed conclusion from the sequencing: three days after Altman endorsed Dario Amodei's call to pace frontier development and pledged the same outside-evaluator access, OpenAI asked investors for a 76% markup — "which tells you the market is treating pacing as a disclosure commitment with no cost to growth."
Complicating the picture is a new antitrust lawsuit filed in the Northern District of California, which accuses OpenAI, Anthropic, SpaceXAI and Google of entering an illegal agreement to slow AI development after their leaders publicly backed Amodei's September essay urging industry-wide cooperation. The plaintiffs — paid subscribers to ChatGPT, Claude, Grok and Gemini — argue the coordination reduced the value they receive. The suit acknowledges that companies may individually choose to slow progress for safety reasons, which is precisely the line the case will turn on.
The bigger risk runs the other way. The longer the IPO is deferred, the more time the market has to answer a question it has so far been willing to postpone: whether the enormous capital being poured into AI infrastructure and software will actually produce returns to match. If the answer disappoints, today's markup will look less like conviction and more like a window that closed.
Comments (0)
Log in to join the discussion
Log InNo comments yet