The internal spreadsheets at OpenAI are currently bleeding $17 billion a year in cash burn, but the primary tension in the C-suite isn't the red ink—it's the calendar.While CEO Sam Altman is reportedly pushing for a landmark IPO as early as the fourth quarter of 2026, his Chief Financial Officer, Sarah Friar, is quietly signaling that the company is nowhere near ready for the scrutiny of the public markets.
OpenAI Wants to Go Public. First Sarah Friar has to transform a sprawling, chaotic research lab into a predictable enterprise machine. It’s a tall order for anyone, but Friar—a veteran who navigated Square’s 2015 debut and steered Nextdoor through the SPAC era—knows that Wall Street doesn't value "vibes" or AGI prophecies. They value gross margins, and OpenAI’s current 33% margin is a far cry from the 70-80% software-as-a-service (SaaS) benchmarks investors expect.
The stakes couldn't be higher. With a staggering $852 billion valuation following a $122 billion funding round in March, OpenAI is effectively a "nation-state" startup. If Friar fails to professionalize the operation, the largest IPO in history could instead become the most expensive cautionary tale in Silicon Valley history.
The $600 Billion Infrastructure Trap
The fundamental disagreement between Altman’s aggressive expansionism and Friar’s financial pragmatism centers on a "poison pill" of infrastructure commitments. Under Altman’s direction, OpenAI has secured a staggering $600 billion in future computing contracts.These are rigid, long-term obligations that don't shrink even if revenue misses its mark—which it already has.
Earlier this week, reports surfaced that OpenAI missed its internal target of 1 billion weekly active users by the end of 2025.In the enterprise segment, the company has begun losing ground to Anthropic, whose Claude 4.5 model has proven more capital-efficient and easier for developers to integrate.
"Sarah Friar has flagged significant concerns about the 2026 IPO timeline, citing organizational unpreparedness and procedural gaps.The internal disagreement reflects a fundamental tension between Altman’s $600 billion spending plan and the reality of slowing revenue growth.Public investors will demand several additional quarters of steady performance to understand how these trillion-dollar infrastructure deals actually generate free cash flow."
— Analysis from PitchBook Intelligence
Global Context: The Race to the Public Markets
OpenAI isn't operating in a vacuum. In London, the Financial Conduct Authority (FCA) has recently tightened listing rules for "high-risk" tech entities, and in Singapore, the Monetary Authority (MAS) has signaled it will require deep audits of AI-generated revenue before allowing local listings.
Closer to home, the competitive landscape is shifting. If Databricks or Anthropic manages to list ahead of OpenAI on cleaner economics, they will set the valuation framework for the entire sector. If OpenAI enters a market it didn't shape, it risks being priced as a capital-intensive infrastructure play (like a utility) rather than a high-margin software leader.
The Financial Crossroads: OpenAI vs. Anthropic (Q1 2026)
Metric | OpenAI | Anthropic |
Annualized Revenue per Employee | $5.6 Million | $6.0 Million |
Gross Margin | 33% | 41% |
Infrastructure Obligation | $600 Billion | $52 Billion |
Weekly Active Users | 910 Million | 140 Million |
OpenAI Wants to Go Public. First Sarah Must Fix the Fundamentals
Friar’s "adult in the room" strategy involves more than just balancing the books. She is reportedly overseeing a massive restructuring of the Microsoft partnership, ending key exclusivity provisions to allow OpenAI to sell its models through Amazon and Google.
This move is a double-edged sword. While it expands the potential customer base, it also strips away the "protected" revenue streams that investors once viewed as a safety net. OpenAI Wants to Go Public. First Sarah Friar has to prove that ChatGPT can stand on its own two feet as a productivity platform—capable of editing spreadsheets and building presentations—rather than just being a clever chatbot that people use until the free credits run out.
Key Takeaways for Founders and Operators
Efficiency over Scale: In 2026, the market is over "growth at any cost." If your revenue per employee isn't climbing alongside your headcount, you aren't ready for an IPO.
The IPO Escape Hatch: Following the successful debut of ARM, the benchmark for tech listings has shifted toward "predictable infrastructure." OpenAI must prove its compute costs are under control.
The "Sarah Friar" Playbook: Founders should look to hire a "Chaperone CFO" early. Professionalizing internal reporting and compliance isn't a "pre-IPO" task; it's a prerequisite for surviving the next decade of AI regulation.
The Perspective Piece: The Mirage of the Trillion-Dollar Debut
There is a dangerous arrogance in assuming that being "first to AGI" translates to being a "good business." We’ve seen this movie before with Uber and WeWork. The "Altman Era" was defined by a scorched-earth approach to data and compute. The "Friar Era" must be defined by the boring, difficult work of corporate governance. If she can’t get the organization to grow up by Q4, the IPO won't just be delayed—it will be a massacre.
What to Watch Next
The civil trial between Sam Altman and Elon Musk, which kicked off on April 29, 2026, adds another layer of complexity. If Musk’s legal team unearths emails suggesting that the transition from non-profit to for-profit was procedurally flawed, it could trigger a "Code Red" for the SEC.
Sarah Friar’s stated target of a 2027 listing seems increasingly like the only sane path forward. As OpenAI Wants to Go Public. First Sarah must convince her CEO that being the biggest company on the planet is meaningless if you can't prove you're built to last. For the global startup ecosystem, the outcome of this internal power struggle will dictate whether the AI boom ends in a soft landing or a generational crash.
