OpenAI has confidentially filed for an initial public offering. The move caps years of breakneck growth and sets the stage for one of the largest market debuts ever. Yet the company’s books tell a story that public investors have rarely encountered at this scale.
Zero debt. Minimal capital spending on the balance sheet. Just $46 million in quarterly capital expenditures as of the first quarter of 2026. That figure trails even Salesforce, a firm that primarily sells software rather than trains frontier models. On paper OpenAI looks like a nimble software operation. Reality paints a heavier picture.
The off-balance-sheet reality now faces regulatory scrutiny
Some $665 billion in future commitments for chips and data centers sit largely off the books, according to The Information. These obligations, tied to deals with Microsoft, Oracle, Amazon, and specialized providers such as FluidStack and the Stargate project, dwarf conventional capital spending. They reflect a deliberate strategy. OpenAI keeps its reported assets light by partnering aggressively instead of owning infrastructure outright.
But that approach carries risks. Public markets demand transparency. Auditors and the Securities and Exchange Commission will examine how these massive future payments appear in financial statements. Investors, too, must weigh whether the structure protects flexibility or simply defers unavoidable costs.
The numbers elsewhere reinforce the tension. OpenAI generated roughly $13.1 billion in revenue last year while incurring total costs and expenses of $34 billion. Losses reached approximately $38.5 billion in 2025, nearly eight times the prior year’s shortfall. Cash burn hit $3.7 billion in the first quarter of 2026 alone. And 72 percent of revenue flows to related parties, chiefly Microsoft.
Such concentration raises eyebrows. Microsoft, a major investor and cloud partner, accounts for the bulk of that figure. The arrangement blurs lines between customer, supplier, and backer. It worked in private markets. Public shareholders may view it differently.
Yet growth remains explosive. Annualized revenue run rates have climbed toward $20 billion or more in recent disclosures. Advertising revenue could reach $102 billion by 2030 under optimistic projections. The company plans heavy infrastructure outlays, telling investors it expects to spend about $600 billion on AI compute by 2030. Those forecasts explain the giant commitments.
Sam Altman, OpenAI’s chief executive, has long argued the firm’s model scales with the value of intelligence itself. In a January 2026 blog post the company stated it manages capacity by “keeping the balance sheet light, partnering rather than owning, and structuring contracts with flexibility across providers and hardware types.” Capital deploys in tranches tied to real demand. The approach lets OpenAI lean forward during surges without overcommitting when growth slows.
That philosophy produced stunning private valuations. OpenAI closed a record $122 billion funding round in March 2026 at an $852 billion post-money valuation. Anchored by Amazon, Nvidia, and SoftBank, the deal ranks as the largest private funding round in history. The valuation jumped nearly tenfold from early 2024 levels around $86 billion.
Now the public test arrives. OpenAI filed its confidential S-1 on June 8, 2026, according to a company blog post. Goldman Sachs and Morgan Stanley are guiding the process. A debut as soon as this fall remains possible though timing could slip. The filing follows Anthropic’s recent submission and precedes expected listings from SpaceX and others. Wall Street confronts a cluster of AI giants seeking capital amid questions over profitability.
Analysts point to steep cash burn projections. Estimates suggest annual losses could climb further before any breakeven, potentially not arriving until 2029 or 2030. One analysis from Sacra cited in recent coverage projects $27 billion in cash burn for 2026. Traditional metrics offer little comfort. Revenue multiples in the private round exceeded 30 times current figures. Public comps for unprofitable tech firms rarely stretch so far.
But. The opportunity appears equally outsized. ChatGPT and successor models command hundreds of millions of weekly users. Enterprise adoption accelerates. New products, including a reported “super app,” aim to broaden the addressable market beyond chat interfaces. If OpenAI captures even a fraction of projected AI infrastructure demand, the economics could improve dramatically over time.
Still, execution challenges mount. The company missed some internal revenue and user targets in its sprint toward an IPO, The Wall Street Journal reported earlier. Training costs for ever-larger models continue to escalate. Competition from Anthropic, Google, and startups intensifies. And regulatory attention on AI safety and energy consumption adds another layer of uncertainty.
Recent coverage highlights the stakes. The Next Web noted on June 22, 2026, that OpenAI’s suspiciously clean balance sheet is about to get a hard look as commitments head for regulators’ desks. The article underscores how the $665 billion in off-book obligations could reshape perceptions once fully disclosed.
Other developments add color. Bloomberg and Yahoo Finance detailed the March funding round’s scale and participants. SmartAsset outlined the valuation trajectory and potential IPO range above $850 billion, possibly stretching toward $1 trillion. Reuters earlier signaled preparations for a listing that could raise $60 billion or more.
So what happens next? IPO prospectuses will reveal audited figures, risk factors, and management’s discussion of liquidity and commitments. Expect extensive footnotes on those $665 billion in obligations. Underwriting banks must convince institutional investors that the model justifies the premium. Retail participation, if allowed, could amplify volatility.
OpenAI’s path differs from past tech IPOs. Unlike early cloud providers that owned their data centers, this firm bets on partnerships and rapid iteration. The bet paid off in private capital. Public markets will render their verdict on whether the light balance sheet signals discipline or deferred reckoning.
One thing appears clear. The AI race has reached public markets. Valuations, cash consumption, and hidden liabilities will face sustained examination. OpenAI’s filing marks not just a corporate milestone but a broader test for an industry built on enormous promises and equally enormous spending.
WebProNews is an iEntry Publication