OpenAI should not rush its IPO just to win the AI race

OpenAI is preparing a giant IPO while still losing billions and lacking durable profits.

OpenAI is not proving that it deserves a public listing so much as proving that capital markets will tolerate almost anything if the growth story is big enough. The company has filed confidentially with the SEC, is targeting a raise that could reach $60 billion, and is carrying a private valuation that already sits near $852 billion. At the same time, internal projections point to roughly $14 billion in losses in 2026 and cumulative losses that may not turn positive until 2029. That is not the profile of a business ready for public discipline; it is the profile of a company trying to buy time with scale.

First argument: the numbers still look like a bet, not a business

Public markets reward companies that can explain how today’s losses become tomorrow’s operating leverage. OpenAI has not done that yet. Revenue has climbed fast, from about $2 billion in annualized revenue at the end of 2023 to more than $20 billion by the end of 2025, but the burn is still enormous. A business that is expected to lose $14 billion in a single year while chasing a $60 billion IPO is asking investors to underwrite future dominance, not current fundamentals.

The valuation gap makes the problem sharper. At an $830 billion valuation, OpenAI’s price-to-sales ratio works out to roughly 65 times 2025 revenue. That is an aggressive multiple even for a category leader, and OpenAI is not operating in a vacuum. Google, Anthropic, and a growing stack of enterprise AI vendors are all pressing on the same market. When a company is priced like a monopoly before it has shown monopoly economics, the public listing becomes a test of investor faith rather than a rational capital event.

Second argument: the structure is finally public-ready, but the economics are not

The 2025 restructuring removed the main legal obstacle to an IPO. OpenAI moved from a nonprofit-led setup into a public benefit corporation for the for-profit arm, while the nonprofit foundation retained a 26% stake worth about $130 billion. That change matters because it clears the path for listing and makes the governance story legible to regulators and investors. It also means the company can now do what it has wanted to do for years: tap public capital at scale.

But legal eligibility is not the same thing as investment readiness. OpenAI’s history as a mission-driven lab is exactly why public investors should be cautious. The company has to balance commercial expansion, safety commitments, and stakeholder obligations under a structure that was built to soften the conflict between mission and profit. That tension does not disappear because the paperwork is complete. If anything, a public listing amplifies it, since every quarter will force OpenAI to defend both its AI roadmap and its margin profile in front of a market that cares mostly about cash flow.

The counter-argument

The strongest case for the IPO is that OpenAI is not a normal software company. It is building the core layer of a new technology stack, and that kind of platform often needs absurd amounts of capital before profits arrive. The company is already operating at internet scale, with ChatGPT still near 900 million weekly active users and enterprise demand expanding across subscriptions and API usage. Supporters also point out that the alternative to public capital is slower infrastructure buildout, weaker competitive positioning, and a real risk that rivals with deeper balance sheets shape the market first.

There is also a strategic argument. OpenAI is tied to a larger AI infrastructure push through Stargate, a $500 billion joint venture aimed at building massive data center capacity. If the company wants to control more of the supply chain, lock in compute, and fund a long runway of model training, public equity gives it a war chest that private markets cannot match. In that view, the IPO is not a vanity event. It is a financing tool for an industrial-scale buildout.

That argument is real, but it does not justify rushing. OpenAI should go public only when it can show a credible path to durable operating discipline, not just a bigger spend plan. The market will fund infrastructure if it trusts the economics. Right now, the company is asking investors to fund growth, compute, and strategic ambition all at once, while admitting that profits may not arrive until 2029. That is too much uncertainty to package as readiness.

What to do with this

If you are an engineer, PM, or founder, treat OpenAI’s IPO as a warning about capital intensity, not a template for success. Build products with a clear unit-economics story, not just usage growth. If your roadmap depends on endless fundraising, you do not have a business model yet. If you are leading a company in AI, use this moment to pressure-test your own burn, pricing, and infrastructure commitments before the market does it for you.