OpenAI hits roughly $2 billion monthly run rate, up about tenfold

OpenAI is starting to look less like a fast-growing startup and more like a full-blown platform company. The big new number is simple — about $2 billion in revenue every month. That is a huge jump in a short time, and it helps explain why investors were willing to fund the company at an $852 billion valuation in a record $122 billion round that closed on March 31. (cnbc.com) ### What actually changed? What changed is that OpenAI stopped talking in vague growth language and gave a very concrete revenue figure. In its March 31 funding announcement window, the company said it was generating $2 billion per month, and CNBC noted that OpenAI made $13.1 billion in revenue in 2025. That means the business is not just growing — it is compounding into a scale that used to be reserved for giant public software companies. (cnbc.com) ### Where is that money coming from? A lot of it still comes from ChatGPT subscriptions, but the mix is broadening. OpenAI said ChatGPT had more than 50 million subscribers as of March 2026, and business usage has become a major revenue engine too. Basically, consumer AI got the company to escape velocity, but enterprise demand is what makes the revenue line look durable instead of faddish. (cnbc.com) ### Why does user scale matter so much? Because AI economics are weird at first. You spend enormous amounts on chips, data centers, and model training before the business looks efficient. The payoff comes when a giant user base starts spreading those fixed costs across billions of prompts. OpenAI said ChatGPT had more than 900 million weekly active users in March, while an earl(cnbc.com)ctive users in 2025. That tells you usage is still climbing fast, not flattening. (cnbc.com) ### So why are investors so excited? Because this is the shape of a winner-take-most market. If one company can attract hundreds of millions of weekly users, convert tens of millions into paying customers, and keep improving product quality, it can justify spending ahead of revenue for a long time. That is the logic behind the $852 billion valuation and the willingness of investors like SoftBank, Amazon, and Nvidia to keep writing giant checks. (cnbc.com) ### Doesn’t AI still cost a fortune to run? Yes — and that is the catch. OpenAI is still not profitable. CNBC said the company is still burning cash even at this new revenue scale. So the headline is not “OpenAI cracked software margins.” The headline is “OpenAI built enough demand that investors believe margins can come later.” That is a very different thing. (cnbc.com)lling inference cost change the story? Because cheaper inference turns usage growth from a problem into a business model. If every answer stays expensive, more users just mean more pain. But if cost per token keeps falling, then each new user becomes easier to serve and easier to monetize. That is why the market keeps treating AI as a future infrastructure layer, (cnbc.com) around $1.339 trillion by 2030. (forbes.com) ### Is this just an OpenAI story? Not really. OpenAI is the clearest proof point, but the bigger story is that AI has crossed from novelty into habit. People use it for work, writing, planning, and everyday questions, and adoption is spreading globally rather than staying concentrated in early-tech circles. Once that happens, revenue can rise much faster than most people expect. (o([forbes.com)## Bottom line? The important number is not just $2 billion a month. It is what that number implies — AI demand is now big enough to support giant recurring revenue businesses even before the economics are fully cleaned up. OpenAI looks like the first company to hit that wall at scale, and everyone else in tech is now racing to prove they can do the same. (cnbc.com)