The company that kicked off the AI gold rush is finally going public — and the numbers say they've already won.
The Summary
- OpenAI's annualized revenue has hit $40 billion, roughly double its end-of-2025 run rate, as it prepares for an IPO
- This makes OpenAI one of the fastest-scaling enterprise software companies in history, reaching revenue heights that took Salesforce nearly two decades
- The IPO timing signals confidence that the company has moved beyond research project into durable commercial infrastructure
The Signal
OpenAI just posted a $40 billion revenue run rate, doubling in roughly eight months. That's not software-as-a-service growth. That's infrastructure replacement velocity. The company that gave us ChatGPT is now pulling in enterprise money at a pace that suggests businesses aren't experimenting anymore. They're rebuilding their operations around AI agents.
The run rate tells you what companies are actually spending, not what they're promising to spend. $40 billion annualized means OpenAI is now closing in on the revenue scale of Salesforce, which took 23 years to reach. OpenAI did it in under four years from commercial launch.
"The run rate doubled in eight months while the company prepared to go public — exactly when you'd expect growth to slow if the foundation was shaky."
Here's what matters: the IPO timing. OpenAI could have stayed private longer. The company had access to capital. But they're choosing to go public now, when they can demonstrate not just revenue growth but revenue acceleration. That's a signal to the public markets that this isn't a demo. It's a platform that enterprises are embedding into their operations.
The breakdown isn't disclosed, but the velocity points to enterprise API revenue, not just ChatGPT subscriptions. Consumer subscription math doesn't get you to $40 billion this fast. What gets you there is companies paying five-figure monthly bills to run agents that replace human workflows. Customer service automation. Code generation. Document processing. The boring, high-volume work that actually pencils out at scale.
Key revenue implications:
- $40B run rate suggests enterprise API is now the majority of revenue, not consumer subscriptions
- ChatGPT Plus at $20/month would need 167 million subscribers to hit this number — OpenAI doesn't have that
- API pricing and enterprise deals are the real engine, meaning OpenAI is now selling picks and shovels to companies building their own agent layers
OpenAI going public also means they're ready to disclose gross margins, burn rate, and compute costs. That's the data the market actually cares about. If they're willing to show that, it means the unit economics work. You don't volunteer to report quarterly earnings if your AI models cost more to run than customers pay to use them.
The IPO also crystallizes the moat question. OpenAI's advantage isn't just the models. It's the API surface area, the developer ecosystem, and the enterprise contracts that are now live and renewing. Once a company rebuilds a workflow around your API, switching costs aren't trivial. That's the difference between a product and a platform.
The Implication
Watch the IPO roadshow. The metrics OpenAI discloses will set the template for how the market values AI infrastructure companies. If gross margins are strong and revenue retention is high, every AI company with enterprise traction will follow them into the public markets. If the numbers are thin, the private valuations across the sector will compress.
For companies building on OpenAI's API: the IPO means the platform is here to stay, but it also means pricing pressure. Public companies have to show margin expansion. That could mean price hikes for API calls, which changes the economics of agent applications built on top. If you're building a business that relies on OpenAI's models, stress-test your unit economics against a 20-30% price increase.