OpenAI CFO signals 2027 IPO delay

OpenAI’s latest drama is not about models. It’s about timing, money, and whether the company can survive public-market scrutiny while still spending like a wartime startup. Sarah Friar, OpenAI’s CFO, is reportedly arguing that an IPO should wait until 2027 rather than happen in late 2026, while Sam Altman has been pushing for a faster path. That matters because OpenAI has already done a lot of the corporate cleanup needed for a listing. The hard part now looks simpler and uglier — can the business justify the bill? ### Why is the date slipping? Because an IPO is not just a paperwork event. Friar’s concern, in the reporting around this fight, is that OpenAI may not be operationally and financially ready for public markets by 2026. Public companies have to on it at the same time. ### Why is Friar the one waving the red flag? That is basically her job. Friar came to OpenAI after running Nextdoor and earlier serving as CFO at Square, so she is the adult in the room for capital structure, reporting discipline, and investor credibility. If a CEO wants maximum speed, a CFO is the one who has to ask whether the numbers survive daylight. At OpenAI, that tension seems to have become the real story. ### Didn’t OpenAI already fix the structure problem? Mostly, yes. OpenAI said its for-profit business had transitioned into a public benefit corporation, with the nonprofit still in control. That matters because a weird old structure was one obvious obstacle to a future IPO. Once that piece moved, the next bottleneck stopped being legal form and became business readiness — monetization constraints and giant capital needs. ### So what is the real argument underneath? Compute. OpenAI is trying to lock in massive infrastructure for training and serving models, and the reported internal debate is over whether the company should commit to that scale before revenue growth proves it can carry the load. A private company can sell the dream. A public company has to defend the quarterly math. That is why monetization is keeping up with ambition. ### Why does public-market timing matter so much? Because IPOs freeze a story into a valuation. If OpenAI goes out too early, investors may focus less on “winner takes most” and more on burn, dilution, and dependence on outside infrastructure gap. ### Is this a sign OpenAI is in trouble? Not exactly. It is a sign that OpenAI is becoming a normal giant company in one very specific way — the finance team is now openly wrestling with the cost of being first. The company just raised $122 billion, which buys time and firepower. But huge funding rounds do not erase the IPO question. They sharpen it, because every new dollar raises the bar for what “ready” has to mean. ### What should readers actually watch now? Watch for three things. First, whether OpenAI keeps talking about a 2026 listing or goes quiet. Second, whether revenue commentary starts sounding steadier than compute commentary. Third, whether the company’s governance and investor messaging look more like a mature platform business and less like a moonshot with receipts so it looks like a plan instead of a delay. ### Bottom line? The interesting part is not that a CFO wants more time. The interesting part is why. OpenAI appears to have moved past the easy question — “can this structure ever go public?” — and landed in the harder one: “can this business go public while spending at AI-war scale?” Friar seems to think 2026 is too soon. If she is right, 2027 is not a miss. It is the first honest date.