- OpenAI missed its internal goal of 1 billion weekly active ChatGPT users by year-end 2025 — a milestone the company still hasn’t announced — and missed its annual ChatGPT revenue target after Google Gemini’s late-year surge ate into market share.
- CFO Sarah Friar has privately told company leaders she is worried OpenAI may not be able to pay for future computing contracts if revenue growth doesn’t accelerate — a significant concern given the company has committed to roughly $600 billion in future data center spending.
- Board directors have been questioning CEO Sam Altman’s push to secure even more computing capacity despite the slowdown, and Friar has expressed reservations about OpenAI’s aggressive IPO timeline, warning that the company’s internal controls aren’t yet ready for public company reporting standards.
- OpenAI’s $122 billion fundraise — the largest in Silicon Valley history — gives it a financial cushion, but the company expects to burn through that sum in three years even if it hits its ambitious revenue targets, some of which are conditional on specific partner agreements.
What Happened?
OpenAI missed a string of internal growth targets over the past several months, according to people familiar with the matter — including a goal of reaching one billion weekly ChatGPT users by the end of 2025, an annual ChatGPT revenue target undercut by Google Gemini’s rapid growth, and multiple monthly revenue targets in early 2026 as Anthropic gained ground in the coding and enterprise markets. The shortfalls have triggered internal tension: CFO Sarah Friar has told company leaders she is concerned about OpenAI’s ability to meet its enormous future compute obligations if growth doesn’t reaccelerate. Board directors have responded by scrutinizing Altman’s data center deals more closely, pushing back on his drive to lock up additional computing capacity. Friar has also told executives and board members that OpenAI isn’t ready for the rigorous reporting standards of a public company, putting her at odds with Altman on IPO timing. Both issued a joint statement insisting they are “totally aligned” on compute strategy and calling any suggestion of division “ridiculous.”
Why It Matters?
For years, OpenAI’s story has been one of seemingly unstoppable growth justifying ever-larger spending commitments. That narrative is now under stress. The company is locked into roughly $600 billion in future data center spending — a sum it committed to when ChatGPT’s dominance seemed unassailable. But competition has arrived faster than expected. Google’s Gemini has taken meaningful share. Anthropic has beaten OpenAI on coding and enterprise benchmarks. Subscriber defection rates are a concern. The $122 billion fundraise buys time, but only if ambitious revenue targets are met — and some of that funding is conditional. An IPO at the end of 2026 would put all of this under public investor scrutiny for the first time, with Friar’s internal controls warning suggesting the company may not be operationally ready for that level of disclosure. The disconnect between Altman’s expansive public posture and the private financial concerns is significant — and this WSJ report makes it public.
What’s Next?
The IPO timeline — currently targeting year-end 2026 — is the most immediate pressure point. Friar’s reservations on readiness, combined with a leadership vacuum from COO Fidji Simo’s unexpected medical leave, add friction to an already complex pre-IPO process. On the revenue side, the growth of Codex and the release of GPT-5.5 (which topped multiple benchmarks) are the most credible near-term catalysts. The compute strategy debate — whether to keep locking up capacity or pull back — will also shape OpenAI’s cost structure and cash burn trajectory in ways that matter enormously to any potential public market investors.
Source: The Wall Street Journal
