Microsoft and OpenAI’s AGI clause is officially dead — and that changes everything

Remember when Microsoft and OpenAI had that weird, semi-exclusive thing going on? The one where Microsoft got first dibs on everything OpenAI built, but only until the mythical AGI arrived? Yeah, that’s over now.

On Monday morning, Microsoft announced a handful of big changes to its long-standing OpenAI deal. The headline: Microsoft will remain OpenAI’s “primary cloud partner, and OpenAI products will ship first on Azure, unless Microsoft cannot and chooses not to support the necessary capabilities.” That’s a lot of qualifiers for a primary partnership.

But the real shift is this: OpenAI can “now serve all its products to customers across any cloud provider.” That means OpenAI is finally free to court enterprise customers on AWS, Google Cloud, or wherever else they want. And the AGI clause that once dictated the future of their deal? Officially gone.

For years, that clause was the elephant in the room. It said that once OpenAI achieved AGI — artificial general intelligence — Microsoft would lose its exclusive license to OpenAI’s technology. It was a neat legal fiction: OpenAI could claim it was building something too powerful to be owned by a single corporation, while Microsoft got to pretend it was just a patient investor. But in practice, it was always going to be a mess.

Now it’s dead. And honestly, I’m not surprised. The AGI clause was always more of a PR shield than a real contractual mechanism. It let OpenAI dodge questions about concentration of power while making Microsoft look like the good guy who would step aside when the big moment came. But as OpenAI’s value skyrocketed and Microsoft deepened its integration, the clause became an obstacle to actual business.

So what does this mean? For one, OpenAI can now sell directly to anyone, on any cloud. That’s huge for their enterprise ambitions. They’ve been trying to break into the corporate market for a while, but being locked into Azure was a hard sell for companies already invested in AWS or GCP. Now they can pitch their models without the cloud baggage.

For Microsoft, it’s a bit more complicated. They still get first crack at OpenAI’s latest models on Azure, but that “unless Microsoft cannot and chooses not to support the necessary capabilities” caveat is a big loophole. If Microsoft drags its feet on some new capability, OpenAI can just go elsewhere. And given how fast the AI landscape moves, that’s a real risk.

But let’s be real: Microsoft isn’t losing sleep over this. They’ve already got Copilot baked into everything from Office to Windows, and they’re investing heavily in their own models. The OpenAI partnership was always a hedge — a way to stay relevant while building in-house. Now they’ve got more flexibility too.

The bigger story is that the AGI clause was always a bit of a fantasy. AGI is still a distant goal, and defining it is a philosophical exercise more than a legal one. By dropping the clause, both companies are admitting that the partnership is just business now. No more grand narratives about saving humanity from runaway AI. Just a straightforward commercial deal.

I’ve been watching this relationship since the early days, and honestly, this feels like the right move. The AGI clause was a nice story, but it was never going to work in practice. Now both companies can focus on what actually matters: building products, selling them, and competing with everyone else.

Will this lead to OpenAI becoming more independent? Probably. Will Microsoft regret it? Only if OpenAI starts winning deals with Microsoft’s biggest competitors. But for now, it’s a clean break from a fiction that outlived its usefulness.