OpenAI, the company behind ChatGPT, confirmed Monday it had confidentially filed preliminary paperwork for a US initial public offering with the Securities and Exchange Commission. It declined to set a timeline, but left open the possibility of a listing as early as this fall.
The company anticipated the filing would leak and chose to announce it first. “It may be a while because there are things we want to do that are likely easier as a private company,” OpenAI said in a statement. “But it’s a complicated set of tradeoffs and this gives us the option to go public sooner if that ends up being best.”
With that announcement, the IPO queue was complete. Rival Anthropic had filed confidentially on June 1. Elon Musk’s SpaceX began trading this week. Together, the three companies are targeting a combined market capitalization approaching $4 trillion — the most consequential test of investor appetite for high-growth technology stocks in a decade. The question worth asking is not whether these companies will eventually go public — they will — but why all three are moving at once, and with such urgency. The answers reveal as much about risk as they do about ambition.
Also read: What Supreme Court’s proposed regulations for AI use in courts mean
The capital logic
The straightforward case for going public is simple: Building artificial intelligence at scale is extraordinarily expensive, and private markets — private equity, sovereign wealth funds and corporate strategic investors — have limits even at unprecedented levels of generosity.
OpenAI told investors earlier this year it plans to spend roughly $600 billion on AI infrastructure by 2030. It has been raising capital at an $852 billion valuation, with annualized revenue crossing $20 billion — yet the company does not expect to turn a profit before 2030. Anthropic, valued at $965 billion after a $65 billion funding round, posted first-quarter revenue of less than $5 billion.
OpenAI’s chief financial officer Sarah Friar told The Associated Press in April that public markets are “much bigger than the private markets,” adding that going public brings a “credentializing moment” — SEC oversight, balance sheet scrutiny, and the discipline of quarterly reporting.
That logic would hold in any market conditions. What makes the current moment distinctive — and what explains the rush — is that conditions favorable to these listings may not hold indefinitely.
Also read: Apple gives developers new keys to its AI future
The window, and the fear of missing it
For the past year, investor appetite for AI stocks has been close to insatiable. Valuations have expanded well ahead of revenues. The AI boom has driven benchmark indices to record highs. That backdrop is precisely what makes a public listing attractive — and precisely what makes timing so critical. The companies racing to file are doing so while the wind is at their backs.
Here, Anthropic’s surprise June 1 filing is an apt illustration of how there is competition within this domain. The filing caught prediction markets off guard — most participants had expected OpenAI to move first. The result: Anthropic’s latest funding round valued it at $965 billion, overtaking OpenAI’s $852 billion valuation for the first time.
As Gil Luria, managing director of DA Davidson, told Reuters: “What OpenAI does not want is for the public market capital to exhaust itself. Not only are SpaceX and Anthropic ahead of it in line to IPO, large public competitors could also raise tens of billions of dollars each in public market secondary issuances, as Google just completed last week.”
Delay too long, and the capital available to welcome you in the equity market shrinks.
There are pressures specific to OpenAI that make the timing still more pointed — the company’s ChatGPT was an early hit, but recently, Anthropic’s Claude has caught up. Emarketer analyst Nate Elliott told AP that the company is at a “precarious moment,” appearing to lose the commanding early leads it built with consumers and businesses to Google and Anthropic. “But OpenAI doesn’t have a lot of other places to look for the enormous capital required to support its costs,” Elliott said.
Also read: Wall Street is rushing to fund the AI bonanza in every conceivable way
The signals turn amber
Here is where the picture grows more complicated. Even as these companies file, warning signs are multiplying that the conditions enabling their sky-high valuations may be shifting.
On Friday, strong US employment data — May nonfarm payrolls came in at 172,000, double consensus expectations — triggered a sharp selloff in technology stocks. The Nasdaq dropped more than 4%, its worst single session since last April’s tariff shock. The SOX semiconductor index fell 10%, its steepest decline since the pandemic, erasing roughly $2 trillion from US equity markets in a single day.
The mechanism, as Reuters markets correspondent Jamie McGeever noted in a column published Monday, is straightforward: a resilient labour market keeps inflation elevated, reduces the likelihood of Federal Reserve rate cuts, and raises the prospect of hikes. A quarter-point rate increase by December is now nearly fully priced by markets. Higher rates raise the cost of capital everywhere, but they are particularly punishing for companies whose valuations rest on earnings projected far into the future.
Goldman Sachs analysts estimate the four largest hyperscalers — Meta, Microsoft, Amazon and Alphabet — will spend $5.3 trillion on AI capital expenditure between 2025 and 2030. That scale of spending raises a question no one has convincingly answered: will the returns justify it?
Citi’s global bear market checklist, which tracks 18 indicators including valuations, fund flows and investor sentiment, is at its most elevated reading since the 2008 financial crisis. Globally, 10 of 18 flags are flashing; on the US checklist, the count stands at 11.5.
Citi noted the tally has not yet reached the “overexuberance” levels that preceded the collapses of 2000 and 2008, but warned that once double digits are breached, deterioration has historically tended to accelerate.
None of this means the AI boom is over, or that these IPOs will fail. Perplexity CEO Aravind Srinivas told CNBC he expects the listings to go well “because they’re doing well,” and SpaceX’s debut this week will serve as the first real stress test of how public markets price this generation of companies.
But the amber signals illuminate why these companies are moving now rather than later.
The capital requirements are structural and permanent. The competitive pressures are intensifying. And the conditions that make a trillion-dollar valuation credible to a public market investor — risk appetite, cheap money, AI euphoria — are not guaranteed to last.
(With inputs from agencies)