Microsoft still gets first shot at shipping OpenAI products on Azure. But OpenAI says it can now serve customers across any cloud provider, Microsoft’s IP license is now non exclusive, and the remaining revenue share now runs one way with a cap through 2030.
Microsoft just gave up some of the contract terms that made it feel like OpenAI's sole landlord. If you build on OpenAI or sell cloud infrastructure, the immediate consequence is not a breakup. It is that OpenAI says it can now serve all its products to customers across any cloud provider, even while Azure keeps first position when Microsoft can support the needed capabilities.OpenAI Microsoft
That is a narrower but still meaningful shift in leverage. Microsoft remains OpenAI's primary cloud partner. OpenAI products still ship first on Azure in most cases. But Microsoft no longer has an exclusive license to OpenAI's models and products through 2032, Microsoft will no longer pay a revenue share to OpenAI, and the revenue-share payments that still run from OpenAI to Microsoft through 2030 now carry a total cap.OpenAI
The money language matters because this deal is easy to misread. OpenAI said Microsoft will no longer pay a revenue share to OpenAI. Separately, OpenAI will keep paying Microsoft a revenue share through 2030 at the same percentage, but subject to a total cap.OpenAI That means one stream of upside to OpenAI disappears, while one stream back to Microsoft stays in place with a ceiling.
The cloud change also needs tighter wording than the first wave of coverage gave it. OpenAI did not say Azure lost first-launch rights or that Microsoft stopped being its main infrastructure partner. OpenAI's statement and Microsoft's matching post both say Azure remains primary. The cleaner claim is that OpenAI says it can now serve customers across any cloud provider without ending Microsoft's priority position.
That distinction matters because parts of this unwind were already public. In an October 2025 partnership update, Microsoft said that OpenAI products outside the API could be served on any cloud provider, while API products developed with third parties would remain exclusive to Azure. In a February 2026 joint statement, OpenAI and Microsoft said the commercial relationship and Azure exclusivity for stateless APIs remained unchanged at that point. Today's amendment looks less like a sudden divorce than another negotiated step away from the original one-cloud structure.
The broader pressure is not hard to see. OpenAI's infrastructure appetite was already straining the idea that one provider could contain the whole relationship. In a November analysis, investor Tomasz Tunguz wrote that OpenAI's compute spending could swell to a scale that makes single-provider dependence hard to sustain. More recently, independent outlet IntuitionLabs argued that OpenAI's reported Oracle arrangement points to a second cloud relationship large enough to challenge any simple Azure-only story. Those are context sources, not proof of today's amendment. They do help explain why exclusive control was getting harder to preserve.
The intellectual-property clause may prove just as important as the cloud one. OpenAI said Microsoft keeps a license to OpenAI intellectual property for models and products through 2032, but that license is now non-exclusive. Microsoft still has access. What changed is that the old bargain, where backing OpenAI also bought the most protected version of that access, is looser than it was.
There is still real counterforce here. The key disclosures came from the companies themselves, so readers can see what the amended terms say, but not what each side had to concede in private to get there. Microsoft also keeps the practical advantage that matters most if Azure can keep up: first position on shipping OpenAI products. What to watch next is whether OpenAI's new freedom shows up in visible multicloud launches, or whether Microsoft's priority rights keep the old hierarchy mostly intact in practice.