OpenAI set aggressive internal targets for revenue and user growth as it barreled toward a public offering. It missed both, according to The Wall Street Journal, which cited people familiar with the company’s internal data. The shortfalls surfaced during the same stretch in which OpenAI was building its IPO narrative, fighting a federal lawsuit from co-founder Elon Musk, and defending a private-market valuation that reached $300 billion after a record $40 billion funding round led by SoftBank in early 2025.
For a company burning through an estimated $5 billion to $7 billion a year on compute, talent, and research, falling behind its own growth benchmarks is not a cosmetic problem. It sharpens a question that underwriters, regulators, and prospective shareholders will need answered before any listing can proceed: Is paying demand for generative AI growing fast enough to justify the price?
Where the growth fell short
The missed targets spanned both revenue and user adoption, the two metrics OpenAI’s leadership had chosen to demonstrate that massive public curiosity about AI was converting into durable, paying customers. The Journal did not specify the exact size of the gap, and OpenAI does not publish audited financials, so outside observers are left to triangulate.
What is known: OpenAI’s annualized revenue reportedly surpassed $5 billion by late 2025, driven by ChatGPT subscriptions, API licensing, and a fast-growing roster of enterprise contracts. The company said in early 2025 that ChatGPT had reached 400 million weekly active users. Those are large numbers by any measure. But hitting a high topline and hitting the steeper trajectory needed to justify a $300 billion valuation are different things. If the adoption curves OpenAI projected internally did not materialize, it suggests that enterprises and consumers are absorbing AI tools more slowly, or more selectively, than the company’s models assumed.
CEO Sam Altman and other executives had not publicly addressed the reported shortfalls as of May 2026. That silence leaves investors to decide for themselves whether leadership views the gap as a temporary lag or a sign that monetizing AI at this scale is harder than the fundraising pitch implied.
The Musk trial adds legal uncertainty
The growth miss landed while OpenAI was simultaneously defending itself in a federal courtroom. Elon Musk, who helped fund the organization’s launch in 2015, took the witness stand to argue that OpenAI’s transformation from a nonprofit research lab into a capped-profit corporation violated the commitments made to him and other early backers. The Associated Press reported that Musk told the judge he felt “genuinely betrayed” by the restructuring, casting it as a departure from the safety-focused, open-research mission he said he was promised.
The lawsuit has a tangled history. Musk originally filed it in February 2024, withdrew it months later, then refiled in August 2024 with broader claims. The case has since escalated into a full trial that puts OpenAI’s governance structure under direct judicial scrutiny. That structure is not an abstract concern. The SoftBank-led funding round was reportedly conditioned on OpenAI completing its conversion to a for-profit entity. A court order altering how the company distributes profits, governs itself, or defines its mission would directly interfere with that conversion and, by extension, the path to an IPO.
Closing arguments are expected in the coming weeks. The range of possible outcomes is wide: the court could side with Musk and force structural changes, reject his claims outright, or impose new governance conditions without dismantling the capped-profit model. Until a ruling is issued, the litigation sits as an unresolved variable in every IPO calculation.
Rivals are not waiting
A revenue miss matters more when competitors are closing the gap. Google has pushed its Gemini models deeper into Search, Cloud, and Workspace, giving enterprise buyers an AI option already bundled with tools they use daily. Anthropic, flush with billions from Amazon and Google, has won enterprise contracts that a year ago would have defaulted to OpenAI. Meta’s open-weight Llama models offer developers a free alternative to OpenAI’s API, lowering the switching cost to zero for price-sensitive teams.
Perhaps most notable is the position of Microsoft, OpenAI’s largest investor and closest distribution partner. Microsoft has continued to integrate OpenAI’s models into its products, but it has also built its own in-house models and opened Azure to rival AI providers. That hedging is rational for Microsoft, but it means OpenAI cannot count on exclusive access to the largest enterprise cloud platform indefinitely.
Every quarter that OpenAI’s revenue or user growth lags its own projections is a quarter in which these rivals gain ground and enterprise procurement teams have more reason to run competitive evaluations before signing long-term AI contracts.
Three hurdles standing between OpenAI and a public offering
OpenAI has not announced a filing date, and no registration statement has appeared with the SEC. CFO Sarah Friar has spoken publicly about the company’s interest in going public, and the nonprofit-to-for-profit conversion Altman has discussed in blog posts and board communications is widely understood as groundwork for a listing. But the combination of missed growth targets and unresolved litigation has complicated the timeline and the terms on which any deal would be struck.
The revenue and user shortfalls are the most immediate obstacle. How OpenAI responds to them, whether through revised guidance, accelerated product launches, pricing changes, or new enterprise partnerships, will signal whether management has a credible plan to close the gap before bankers finalize offering documents. A company that acknowledges a miss and shows a clear correction earns more credibility with institutional investors than one that stays silent and hopes the numbers improve on their own.
The Musk trial verdict is the second. A ruling that leaves OpenAI’s capped-profit structure intact removes a major overhang; one that mandates governance changes or restricts the for-profit conversion introduces new variables that would need to be disclosed in any S-1 filing and priced into the offering.
The third is investor appetite itself. Underwriters will gauge whether large institutional buyers still view OpenAI’s growth trajectory as strong enough to absorb a valuation north of $300 billion in public markets, where scrutiny is sharper and liquidity cuts both ways. Early indications from secondary-market pricing and analyst commentary will shape that assessment well before a roadshow begins.
None of these hurdles is insurmountable on its own. Companies routinely miss internal targets and still go public. High-profile litigation often resolves without derailing corporate strategy. But all three arriving at once, during a period when investors and analysts are already questioning whether AI spending is producing real returns, makes OpenAI’s road to the public markets considerably rougher than it looked even a few months ago. For anyone tracking the AI industry or weighing exposure to OpenAI through secondary shares or a future offering, the next several weeks will carry outsized weight.
