OpenAI is in talks at a $1.2 trillion valuation, days after calling for an AI slowdown

OpenAI is in early talks with investors about a funding round that would value it at more than $1.2 trillion, the Financial Times reported on Tuesday evening, three days after Sam Altman said the company would not go public this year because of the safety work in front of it. Investors started the discussions, the FT said. The valuation would be about 41% above the $852 billion OpenAI carried after closing a $122 billion round on March 31, and 64% above the $730 billion it was worth in February.

The week's argument has been whether the AI industry should slow down. The answer arriving from the capital markets is that the money is not slowing down at all. Anthropic's chief executive published an essay on Saturday asking labs to pace the frontier. Altman agreed, and shelved an initial public offering that had been expected to value OpenAI at $1 trillion or more. Three days later, the private market is discussing a number higher than the one the IPO was supposed to deliver.

What the FT reported about OpenAI's $1.2 trillion round

The talks are preliminary, and it is investors doing the asking. The FT reported that OpenAI is weighing a funding round at a $1.2 trillion valuation ahead of an IPO, and that investors kicked off the discussions, as summarized by Finimize. No round size, lead investor or closing date has been reported. Bloomberg and The Information carried the same valuation figure within the hour.

Set against what OpenAI has already raised, the step-up is fast rather than unprecedented. The company finalized $110 billion at a $730 billion valuation in February, then closed $122 billion at $852 billion on March 31, the largest funding round on record at the time, according to CNBC. Amazon put in $50 billion, with $35 billion of it contingent on an IPO or on reaching artificial general intelligence, while Nvidia and SoftBank each committed $30 billion and Microsoft continued to participate. A $1.2 trillion mark would add roughly $350 billion of paper value in under six months.

Why a private round instead of the IPO

Because the IPO is not happening this year, and OpenAI still needs cash. Altman said on Sept. 12 that going public in 2026 would be "ill-advised," and that the company needed to focus on "meeting this moment of what is going to be required for safety and alignment, and how the industry and governments can work together," Axios reported. OpenAI filed confidentially with the Securities and Exchange Commission in June and chief financial officer Sarah Friar has pointed to 2027 for a listing. TECHi tracks the filing status and timeline on its OpenAI IPO page.

A private round solves the timing problem without the disclosure. It does not require a prospectus, audited quarterly results, a roadmap for safety spending or a risk-factor section describing what happens if governments act on the warnings the company's own chief executive endorsed last weekend. It prices the company among a small group of investors who already own it, rather than in a public market that spent Monday selling anything connected to AI capacity.

The contrast with Anthropic sharpens the point. The company that called for the slowdown is the one heading for public markets: Anthropic is expected to begin marketing an IPO in mid-October at the earliest and to list days before the November midterm elections, with some investors talking about a $2 trillion listing against a last private valuation near $965 billion, Reuters reported via CNBC. TECHi reported last week that Anthropic expects a second profitable quarter as that process moves. One lab is asking public investors to fund a slower frontier. The other is asking private ones to fund the same frontier, faster, and is being offered more money for it.

The slowdown week the money ignored

Look at what else happened in the days around this valuation. Dario Amodei's essay asked for embedded third-party evaluators, coordination among democratic labs and a narrow antitrust waiver so competitors could agree on standards without breaking the law. Altman, Elon Musk and Google's AI chief broadly agreed with the direction. The president called the warnings a hoax and phoned Jensen Huang on stage to say so, which TECHi covered as the day the labs wrote their own rules while Washington declined to write any.

On Tuesday the disagreement moved inside the industry. Huang called Anthropic's proposed antitrust waiver "completely unnecessary" and described the choice between moving fast and pacing the frontier as a false one, CNBC reported from Dreamforce. David Sacks, who co-chairs the president's science and technology council, has made a similar argument: any lab that thinks it should slow down can simply slow down, no waiver required. Both positions share an assumption that a pause is a choice each company makes alone.

That is where the funding round matters. A lab that raises new billions at a $1.2 trillion valuation has made a promise about growth to the people writing the checks. TECHi's read on Monday was that the slowdown has no enforcer: no regulator, no treaty, no binding pledge. A capital structure is the closest thing to an enforcement mechanism in this story, and it points the other way. Investors do not underwrite a 41% step-up in five months to buy a slower company.

What $1.2 trillion implies about OpenAI's revenue

The valuation is easier to judge against sales than against the last round. OpenAI is generating about $2 billion of revenue a month, according to reporting on its finances this year, which annualizes to roughly $24 billion. A $1.2 trillion valuation would therefore price the company at about 50 times annualized revenue, by TECHi's arithmetic on that run rate.

That multiple is not unheard of for a company growing this fast, but it is far above anything in public markets at this size. Nvidia, the most valuable company in the AI complex, trades at a small fraction of that multiple while earning tens of billions of dollars in profit. Nvidia closed Tuesday at $212.17, up 0.6%, per Nasdaq.com.

The gap is the reason the IPO timing matters. Private marks are set by a handful of investors who can accept a story about 2030; public markets reprice that story every day, and this week they repriced the whole AI complex twice, first on the slowdown calls and then on a 10-year Treasury yield above 5%. Staying private at $1.2 trillion lets OpenAI raise at a number a public market might not give it in the same week.

Who is writing the checks, and what they get back

The names in OpenAI's last round explain part of the enthusiasm. Nvidia committed $30 billion to a customer that spends much of it on Nvidia chips. Amazon committed $50 billion to a company that rents cloud capacity, with most of that contingent on an IPO or on AGI. SoftBank added $30 billion, and Microsoft, OpenAI's largest early backer and infrastructure partner, kept participating. These are suppliers and partners financing demand for their own products, a loop TECHi examined when Nvidia became its customers' credit line.

The loop is why the AI trade moves together and why it sold off together on Monday, when the semiconductor index fell 5.9% on the slowdown calls. It also explains why Tuesday's reaction in chip stocks was muted rather than celebratory: Nvidia closed up 0.6% at $212.17, per Nasdaq.com, in a session where the S&P 500 fell 0.45%. A higher private valuation for OpenAI is good for the companies that own a piece of it, but it is not an order for more chips by itself.

For the rest of the market, the number sets a benchmark. Anthropic's bankers will market an IPO in the weeks after this report, into a public market where the 10-year Treasury yield just touched its highest level since 2007 and the Federal Reserve is expected to raise rates on Wednesday. A $1.2 trillion private mark for OpenAI is the comparison every Anthropic investor will use, and a $2 trillion aspiration is the number Anthropic's own backers have floated. Both figures now have to survive a rate environment that did not exist when this cycle of AI funding started.

What happens next to the $1.2 trillion mark

Three things will show whether the $1.2 trillion figure holds. The first is the round itself: size, lead investor and whether any of it is contingent, as $35 billion of Amazon's commitment was. The second is Anthropic's prospectus, expected late this month, which will put audited numbers from a frontier lab in front of public investors for the first time and give the market something other than private marks to price. The third is what OpenAI says about safety spending as it raises. On Tuesday it backed bills in Congress addressing AI and biological-weapon threats, Reuters reported, and it has endorsed a bipartisan House plan for third-party safety assessments, according to Politico. A funding round is the moment those commitments acquire a cost, and investors will want to know what it is.

OpenAI has not confirmed the talks publicly. Early-stage discussions at this scale routinely change price, or fail to close at all, before any money moves.