OpenAI's $1 Trillion IPO Dream Is Fracturing at the Top

Less than three years after the boardroom coup that nearly destroyed it, OpenAI is facing a new governance crisis—quieter than the last but potentially just as consequential. According to a report from The Information, a significant rift has opened between CEO Sam Altman and Chief Financial Officer Sarah Friar over the company’s planned IPO, with the two executives holding fundamentally different views on whether OpenAI is ready to go public by the end of 2026.

The stakes could not be higher. Altman is pushing for a Q4 2026 listing at a valuation of approximately $1 trillion—a figure that would shatter every prior record for a tech IPO. Friar, by contrast, has reportedly told colleagues in private conversations that the company simply isn’t ready, raising concerns about cash burn, the pace of spending commitments, and whether OpenAI’s revenue trajectory can sustain the narrative required to price at that level.

The $600 Billion Problem

At the heart of the disagreement is scale. Altman has publicly committed OpenAI to spending $600 billion over five years, a number that encompasses data center buildout, compute procurement, research headcount, and the infrastructure required to support the company’s rapidly expanding product suite. It is an audacious figure—greater than the GDP of most nations—and Friar is said to have serious reservations about whether projected revenue can justify it.

The numbers are stark. OpenAI’s 2026 operating losses are estimated at roughly $14 billion, even as the company approaches $3 billion in annualized revenue. The company is expected to burn through more than $200 billion before reaching sustained positive cash flow. In the normal cadence of startup growth, losses at this scale would not be disqualifying—but OpenAI is not pursuing a normal growth cadence. It is attempting to price a public offering at $1 trillion, which requires a growth story that can absorb that level of scrutiny from institutional investors who have spent the past two years watching AI valuations fluctuate wildly.

Altman’s bet is that OpenAI’s revenue trajectory—reportedly growing at more than 100% year-over-year—will be compelling enough to carry the deal. Friar’s concern is that the spending commitments make that story harder to tell with a straight face.

A Structural Marginalization

The reported tensions have manifested in ways that go beyond strategic disagreement. According to people familiar with the situation, Altman has been excluding Friar from key investor meetings in the lead-up to IPO roadshow preparations—a significant snub for a chief financial officer whose core job function is managing the company’s relationship with capital markets. Friar now reports to Fidji Simo, OpenAI’s CEO of Applications, rather than directly to Altman—an organizational arrangement that effectively reduces her standing in the company’s most senior conversations.

The dynamic is reminiscent, in spirit if not in mechanism, of the November 2023 board crisis that saw Altman briefly ousted before being reinstated days later. In that episode, a mismatch between Altman’s vision for OpenAI’s commercial expansion and the board’s conception of its nonprofit mission created an irreconcilable conflict. The current tension is more about financial prudence versus ambition, but the underlying pattern—Altman moving faster than internal counterweights are comfortable with—is structurally similar.

Microsoft as the Wild Card

One factor that could shape the IPO outcome is Microsoft, which has invested more than $13 billion in OpenAI and holds a complex set of licensing rights to the company’s models. Bloomberg has noted that Microsoft could function as a kind of anchor investor in an OpenAI IPO, providing institutional credibility and a built-in demand signal that would help set the price at the high end of any range.

The relationship between the two companies has evolved considerably since their initial partnership. Microsoft has built its own internal AI research division under the MAI (Microsoft AI) branding, creating at least a nominal hedge against its dependence on OpenAI. Whether that tension creates complications for Microsoft’s role as IPO stabilizer, or whether financial self-interest overrides everything else, remains one of the key variables in the deal’s construction.

The Governance Question Markets Will Ask

Any OpenAI IPO prospectus will face intense scrutiny of governance arrangements that remain unusual by the standards of public company law. OpenAI’s corporate structure still retains elements of its original nonprofit mission: the for-profit entity is technically “capped” in its returns to investors, and the board retains unusual powers relative to standard Delaware corporate governance. The 2023 crisis revealed how those arrangements can produce unexpected instability.

Institutional investors doing due diligence will ask hard questions: Who has final authority? What checks exist on the CEO? How does the board function in practice? For a company with a history as turbulent as OpenAI’s, those questions will receive additional scrutiny.

The CFO-CEO divide, if unresolved by the time the S-1 is filed, would itself become a disclosure issue. Public companies are required to describe material disagreements among senior executives. A documented fracture between the chief executive and chief financial officer over the company’s core financial strategy would be a significant red flag for a deal priced at any multiple of current revenues—let alone at $1 trillion.

What Comes Next

Neither Altman nor Friar has commented publicly on the reported tensions. OpenAI has not confirmed an IPO timeline. The company’s $122 billion funding round last year—the largest ever for a private company—established the $852 billion valuation that Altman is attempting to surpass at listing.

For Altman, the IPO is not just a financial event. It is a statement about the permanence of OpenAI’s position at the top of the AI industry, a moment of institutional legitimacy that would transform the company from a well-funded startup into a durable public institution. For Friar, the question is simpler and more technical: can the numbers actually support it?

In Silicon Valley, those two questions are rarely resolved by whoever is right. They are resolved by whoever controls the timeline. Right now, it is not entirely clear who that is.