OpenAI's chief financial officer Sarah Friar stepped in front of CNBC cameras last Thursday to lay out what has become one of the most watched financial stories of 2026: a company with $25 billion in annualized revenue, 900 million weekly active users, and an $852 billion post-money valuation that still expects to lose $14 billion this year alone. The informal conversations with Wall Street banks about a fourth-quarter 2026 listing are now open knowledge, and the question investors must grapple with is not whether OpenAI will go public, but on what terms — and at what risk.
The revenue trajectory is genuinely extraordinary. OpenAI crossed $6 billion in annualized revenue at the end of 2024; by February 2026 it had reached $25 billion — a roughly 4x increase in 14 months. That growth rate outpaces Salesforce, Snowflake, and virtually every enterprise software company at comparable scale. The company raised $122 billion in February, closing at an $840 billion pre-money valuation, and has since been hiring aggressively for IPO-readiness roles — a new chief accounting officer and a business finance officer to lead investor relations. All the choreography of a public offering is visibly underway.
"Serving enterprises is a very profitable business at scale, and that's how we will build a sustainable business model." — Sarah Friar, CFO, OpenAI
The problem — and it is a significant one — is the distance between that vision and today's financials. OpenAI does not expect to reach breakeven until 2030 at the earliest. HSBC analysts are more skeptical, projecting that profitability may remain elusive even by 2030 and that the company will require over $207 billion in additional funding through that year. The infrastructure costs underlying this equation are staggering: OpenAI has committed more than $600 billion toward cloud server capacity over the next five years alone, and its cash burn currently exceeds $150 million every single day. Revenue is climbing, but compute costs are climbing alongside it, and the unit economics of consumer AI remain deeply negative — the company is currently gross-margin positive only on its large enterprise contracts, while continuing to lose money on every individual developer and consumer subscription.
The Retail Investor Equation
The decision to reserve approximately 30% of IPO shares for retail investors is one of the more strategically intriguing elements of OpenAI's listing plans. Friar has framed it as democratization — and there is genuine populist appeal to allowing individual investors into what will likely be the largest AI IPO in history. But the allocation also functions as a signal about institutional appetite. OpenAI's $1 billion private placement for retail investors (routed through JP Morgan, Morgan Stanley, and Goldman Sachs) was oversubscribed three times over, suggesting retail demand is real. Whether that demand reflects informed confidence in the business model or a reflexive desire to own a piece of a culturally dominant brand is a distinction that will matter enormously once quarterly losses become public record. The CFO herself has reportedly warned colleagues internally that the company may not be "IPO-ready" by 2026 — a candid acknowledgment that the governance transition from nonprofit to for-profit corporation is still being resolved, and that $600 billion in multi-year compute commitments makes the financial picture unusually complex to present to public-market investors.
The 2026 IPO window is further complicated by supply-side pressure. SpaceX, Anthropic, and OpenAI are all approaching public markets within overlapping timeframes, and analysts have begun flagging the risk of AI IPO demand crowding — three multi-hundred-billion-dollar offerings competing for the same institutional and retail capital in a single calendar year would be historically unprecedented. OpenAI's core strategic argument is that its revenue growth curve will eventually outrun its cost curve. The signal to watch when the S-1 arrives in Q3 is the gross margin trajectory broken down by product segment. If enterprise gross margins are trending toward 60–70%, the long-term profitability case becomes structurally defensible. If costs continue to widen across all segments, the $1 trillion target valuation will be a difficult number to sustain in a public market that ultimately prices on cash generation rather than possibility.