Microsoft and OpenAI’s AGI clause is officially dead. Good riddance.

If you’ve been following the Microsoft-OpenAI saga, you know their relationship has gone from cozy partnership to something more like a situationship—lots of public hand-holding but no real commitment. Well, that just got even less committed.

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 still ship first on Azure—unless Microsoft “cannot and chooses not to support the necessary capabilities.” That last bit is new, and it’s a pretty big escape hatch.

But the real news is that OpenAI can now serve all its products to customers across any cloud provider. That’s a complete reversal from the earlier exclusivity arrangement where Azure was the only game in town for OpenAI’s enterprise customers. And it comes with a quiet but significant death: the famed AGI clause is gone.

For years, that clause was the weirdest part of their deal. It basically said that once OpenAI achieved artificial general intelligence—whatever that means—Microsoft’s access to OpenAI’s tech would be cut off. The idea was that AGI shouldn’t be locked into a single corporate cloud, but in practice it was a ticking time bomb that made long-term planning impossible for both sides.

Honestly, I never understood how that clause was supposed to work in the real world. AGI is a moving target. Sam Altman himself has said it could be here in a few years or decades away. Keeping a contract clause tied to a fuzzy, unverifiable milestone was always going to be a mess. Dropping it was the smart move.

What does this mean in practice? For one, OpenAI can now chase enterprise customers running on AWS, Google Cloud, or whoever else. That’s a big deal for companies that have multi-cloud strategies and don’t want to be locked into Azure just to use GPT-4 or whatever comes next.

For Microsoft, it’s a bit of a loss of leverage, but they still get first dibs on new models and the Azure relationship is still tight. The “unless Microsoft cannot or chooses not to support” clause gives them an out if something goes sideways with OpenAI’s infrastructure demands.

I’ve seen this pattern before. Big tech partnerships start with exclusivity, then slowly unravel as the smaller partner gains power. OpenAI is no longer the scrappy startup that needed Microsoft’s cloud to survive. They’re a giant in their own right. This deal change reflects that reality.

The AGI clause was always more of a philosophical statement than a practical contract term. Killing it removes a distraction and lets both companies focus on what actually matters: making money and building products people will pay for. That’s fine by me.

One thing I haven’t seen many people mention: this also clears the path for OpenAI to eventually build its own cloud infrastructure. They’ve been hiring hardware engineers and dropping hints. If they can sell across any cloud now, they don’t need Azure as a crutch forever. Microsoft knows this, which is probably why they added the “unless Microsoft cannot support” language—it’s their own escape hatch if OpenAI starts competing on cloud services directly.

The bottom line: the Microsoft-OpenAI partnership is now more pragmatic and less theatrical. The AGI clause was a nice story for press releases, but it never made sense as a business term. Good riddance.