OpenAI Plans 2027 IPO as Anthropic Rivalry Intensifies

There's a particular kind of corporate communication that functions less as information and more as morale management. When OpenAI CFO Sarah Friar told employees at an all-hands meeting this week that the company "will be a public company in 2027" — or sooner, if growth holds — she was doing both at once.

The IPO announcement itself isn't a surprise. What's worth examining is why it was delivered the way it was, what the financial picture actually looks like underneath the confidence, and what going public would genuinely mean for a company that has always insisted it isn't quite like other companies.

Running Its Own Race (Or Saying It Is)

The competitive subtext of Friar's remarks was explicit enough that it barely counts as subtext. According to CNBC, she told staff: "As you know we are confidentially under file, and Anthropic is also under file. There is a chance they pull the cover off that confidential file in the coming weeks and become public in September. That's OK, we are running our own race."

"That's OK" is doing a lot of work in that sentence. It's the kind of reassurance you offer when you're fairly sure people are worried it's not OK.

The Tech Buzz and Stocktwits both note that Friar's message was explicitly framed as a dismissal of concern over Anthropic's timeline. Briefs.co flags the "faster if growth holds" qualifier — a hedge worth noting, since it implies the 2027 target is conditional rather than fixed.

The competitive dynamic here matters because IPO timing in a nascent sector is rarely pure finance. Being first to market shapes how public investors mentally categorize you: Are you the category leader or the fast-follower? For AI companies still fighting over that identity, the sequence of public listings could influence the narrative as much as the underlying fundamentals.

The CFO's Job

Sarah Friar's presence at OpenAI is itself part of the story. She joined from Nextdoor, where she had served as CEO, and before that was CFO at Square during its own IPO. Hiring her was a signal — the kind companies send when they're serious about public markets rather than just talking about them. CFOs who've navigated listings before understand how different the discipline is from private-market fundraising, where you can shape your own story. Public markets have analysts, short sellers, and quarterly earnings calls. You can't just reassure staff; you have to produce numbers.

And the numbers, at the moment, are complicated. OpenAI's revenue has climbed sharply — hdfcsky.com notes the company has seen strong revenue growth alongside intensifying competition. But Heise Online reports that losses are widening faster than revenue is growing — a pattern that tends to look inspiring in a bull market and alarming in a correction. The New York Times reported in March 2025 that OpenAI closed a funding round that valued the company at $300 billion — a figure that sets a daunting pricing reference point for any eventual public offering.

Price a listing that honors that valuation and you're asking public-market investors to make a bet that private investors made at the peak of AI enthusiasm. Underprice it and you risk signaling weakness and burning existing investors. This is the tightrope Friar will be walking, and it explains why "2027 or sooner if growth holds" is the framing: it buys time to see which direction the numbers move.

What Going Public Actually Does to an AI Company

The standard case for an IPO is straightforward: access to capital, liquidity for early investors and employees, public profile. For OpenAI specifically, the calculus is more textured.

OpenAI's foundational structure — born as a nonprofit, then restructured into a "capped-profit" model — has always created a complicated relationship between its stated mission (beneficial AGI for humanity) and its capital needs (extraordinary). It has been in the middle of converting to a more conventional for-profit structure, a transition that has attracted its own share of controversy and legal scrutiny. Going public accelerates the finalization of that transformation in a way that's hard to walk back.

Public companies face a specific kind of pressure: quarterly. Biotech is instructive here, and not because the analogy is flattering. Drug companies burning cash on decade-long trials have learned — sometimes painfully — that public investors can handle patience, but only when you give them a believable story about what the waiting is for. The ones that survive are the ones who can explain their burn rate as investment, not failure. OpenAI will need to make that case constantly, to analysts who will ask the same questions every ninety days regardless of what the research team is building. The discipline that requires is genuinely different from managing a cap table of sophisticated venture investors who opted into a long-horizon bet.

The Verge has been tracking another dimension of this: OpenAI has recently signaled a more cautious approach to deployment pace, what The Verge frames as the company "hitting the brakes" on voluntary AI development speed. Whether that's genuine strategic recalibration or reputational management ahead of a listing is, frankly, not something anyone outside the company can answer with certainty. But it's worth holding in mind: companies entering public scrutiny tend to sand their edges. The version of OpenAI that files its S-1 will be presenting itself, carefully.

The Governance Question Nobody's Asking Loudly

Here's what I find most interesting about the IPO trajectory, and what gets the least attention in coverage focused on the competitive horse race: what happens to OpenAI's stated mission once it has a fiduciary duty to public shareholders?

The nonprofit board that fired and then rehired Sam Altman in 2023 was exercising exactly the kind of oversight the original structure was designed to enable. A publicly traded OpenAI will have a different board, different legal obligations, and different stakeholders applying different kinds of pressure. The "don't worry about Anthropic, we're running our own race" message to employees is one kind of reassurance. The message that will need to be delivered to public markets — about growth, margins, competitive moat, and return on capital — is a different kind of pressure entirely.

That's not an argument against going public. It's an observation about what public markets do: they're not passive recipients of corporate vision. They reshape it. The question isn't whether OpenAI can handle an IPO. It's whether the version of OpenAI that emerges on the other side of one will still be running the same race it thinks it's running now.