The $852 Billion Paradox: Inside OpenAI's Race to IPO While Burning $14 Billion a Year

The most valuable private technology company in history has never turned a profit. That paradox sits at the center of what may become the defining corporate saga of the AI era.

OpenAI closed its latest funding round on March 31, 2026 — a record-shattering $122 billion raise that valued the company at $852 billion post-money. Retail investors piled in alongside SoftBank, Nvidia, Microsoft, and Amazon. The round included $3 billion from individual investors via platforms like E*TRADE, marking the first time ordinary Americans could buy a piece of the AGI dream before an IPO.

Two days later, Anthropic reportedly hit a $1 trillion valuation, overtaking OpenAI in the market-cap race for the first time.

And on Monday — April 27 — jury selection begins in Musk v. OpenAI, the lawsuit that could, in theory, unwind the entire for-profit conversion and return the company to its nonprofit origins.

This is the story of a company sprinting toward an IPO while burning $14 billion a year, fighting an existential lawsuit, and watching its closest rival steal the enterprise market from under its feet.

The Numbers Don't Add Up — Until They Do

OpenAI's financials are a study in contradictions.

On the revenue side, the trajectory is staggering. The company hit $25 billion in annualized revenue by March 2026, up from $13.1 billion for all of 2025 and just $3.7 billion in 2024. ChatGPT now has over 900 million users. Enterprise accounts for more than 40% of revenue, with over one million business customers and seven million "ChatGPT for Work" seats deployed across organizations like Morgan Stanley, Cisco, Amgen, and Target.

On the cost side, the picture is grimmer. Internal projections show a $14 billion net loss for 2026 — nearly triple the previous year's losses. Cash burn could hit $17 billion. The company doesn't expect to reach profitability until 2029 at the earliest.

The gap between explosive revenue growth and accelerating losses comes down to one thing: compute. OpenAI is spending money on AI infrastructure at a pace that makes the hyperscalers look restrained. Through Stargate LLC — its joint venture with SoftBank, Oracle, and Abu Dhabi's MGX — the company has committed to $500 billion in U.S. data center spending by 2029, with an initial $100 billion deployment already underway in Abilene, Texas. International buildouts are planned across the UAE, Argentina, and beyond, though projects in the UK and Norway were paused in April due to energy costs and regulatory friction.

In February, OpenAI quietly reset its total infrastructure spending target to $600 billion cumulative by 2030 — down from the jaw-dropping $1.4 trillion figure floated in late 2025. The revision signals pragmatism, but $600 billion is still a number that would make most sovereign wealth funds blink.

The Enterprise War OpenAI Is Losing

The real threat to OpenAI's long-term value proposition isn't the burn rate. It's what's happening in the enterprise market.

Anthropic's Claude has surged past OpenAI in capturing new enterprise AI spending. According to Axios, Anthropic now commands 37-40% of the enterprise LLM market in Q1 2026 — compared to OpenAI's 33%. More strikingly, Anthropic captures 73% of first-time enterprise AI buyers and holds a dominant 54% share in coding-related AI applications.

The numbers tell a story of momentum. Anthropic's annualized revenue hit $30 billion by April 2026, up from approximately $6 billion in 2025 — a fivefold increase. The company now counts over 1,000 enterprise customers spending more than $1 million annually each, with 300,000 business customers in total.

Google's Gemini holds a steady 20-21% share, with 8 million enterprise seats across 2,800 companies and deep integration advantages through Google Cloud and Workspace.

OpenAI's response has been the Frontier platform — an enterprise tool for building and deploying "AI coworkers" with customers like HP, Intuit, Oracle, and Uber — plus updates to its Agents SDK. But there's a structural vulnerability: 95% of ChatGPT's 900 million users pay nothing. Converting free users into revenue while simultaneously losing the enterprise growth race to Anthropic is the central strategic challenge of the pre-IPO period.

This is where the analysis gets actionable. AlphaBriefing members get the full investment framework — scenarios, positioning, and the bottom line.

Subscribe to AlphaBriefing — Free, Member, and Paid tiers available.