OpenAI’s numbers tell two very different stories at once. The company has blown past $25 billion in annualized revenue and is taking early steps toward a public listing — yet it is also burning cash at an extraordinary rate, with internal projections pointing to $14 billion in losses and a deeply negative operating margin. As it eyes an IPO, OpenAI embodies the AI era’s central tension: explosive growth, staggering costs.
The revenue rocket
Growth has been phenomenal. OpenAI surpassing $25 billion in annualized revenue, reportedly more than $2 billion a month, marks one of the fastest ascents in tech history. Demand for ChatGPT, its API and enterprise products has scaled rapidly, validating the commercial appetite for generative AI. On the top line, few companies have ever grown this fast.
The cash bonfire
But the bottom line is brutal. Internal documents reportedly project around $14 billion in losses across 2023-2029, and Q1 2026 reporting implied a roughly negative 122% operating margin — meaning OpenAI spends far more than it earns. The culprit is the colossal cost of compute: training and running frontier models, plus the data-center and chip commitments to support them, dwarf even billions in revenue.
The IPO question
A listing looms despite the losses. OpenAI is reportedly preparing for a public offering, potentially as soon as late 2026 — a move that would test whether public investors will fund a company growing explosively while losing billions. It is a high-stakes bet that the market believes in the long-term payoff enough to look past near-term cash burn.
Why the losses are so big
Frontier AI is ruinously expensive. The compute required to stay at the cutting edge — vast GPU fleets, gigawatt-scale data centers, custom chips — consumes capital faster than revenue can replace it. OpenAI’s parallel efforts to build its own chips and secure enormous data-center capacity are attempts to control those costs, but they require huge upfront spending of their own. The economics of leadership are punishing.
The broader signal
OpenAI is the bellwether for AI’s economics. Its combination of soaring revenue and massive losses crystallizes the question hanging over the whole industry: can the AI boom translate into sustainable profit, or is it growth bought at unsustainable cost? With enterprises trimming budgets and price competition rising, the pressure to prove a path to profitability is mounting — and an IPO would put OpenAI’s finances under a harsh public spotlight.
The bottom line
OpenAI’s $25 billion revenue alongside $14 billion in projected losses captures the paradox of the AI era — extraordinary growth funded by extraordinary spending. As it steps toward a possible late-2026 IPO, the company must convince investors that today’s cash bonfire builds tomorrow’s dominance. Whether the market buys that story will be one of the defining financial questions of the AI age.