What Changed
The core shift is the removal of exclusivity. Microsoft's license to OpenAI's intellectual property and models extends through 2032, but it is now explicitly non-exclusive. OpenAI products will still ship first on Azure unless Microsoft decides otherwise or is unable to support the required capability, but from there they can go anywhere.
The revenue structure has also been revised. OpenAI will continue paying Microsoft a 20% revenue share through 2030, subject to a total cap that has not been disclosed publicly. Critically, that payment is now described as "independent of OpenAI's technology progress" — language that effectively neutralizes the original agreement's AGI clause, which would have triggered a renegotiation of terms if and when OpenAI reached artificial general intelligence. Microsoft, for its part, will stop paying its own revenue share back to OpenAI.
Microsoft retains an equity stake in OpenAI's public benefit corporation valued at approximately $135 billion, representing roughly 27% of the company on a diluted basis. That ownership position is unchanged.
The Context Behind the Move
The signs of strain had been visible for some time. In a memo sent to OpenAI staff earlier this month, Chief Revenue Officer Denise Dresser wrote that the Microsoft partnership had "limited our ability to meet enterprises where they are — for many that's Bedrock." She described enterprise interest in running OpenAI models through AWS as "frankly staggering."
The renegotiation follows a period of rapid expansion in OpenAI's commercial footprint. The company's $110 billion funding round in February — backed by Amazon, Nvidia, and SoftBank — significantly shifted the balance of leverage in its relationship with Microsoft. Where the original deal was struck when OpenAI needed Azure credits to operate, OpenAI now has the financial independence and multi-cloud infrastructure commitments to dictate different terms.
Amazon CEO Andy Jassy confirmed on X on Monday that AWS will make OpenAI models available directly to customers through the Bedrock service within weeks. "With this, builders will have even more choice to pick the right model for the right job," he wrote.
What It Means for Enterprise AI
The practical effect for enterprise customers is greater flexibility. Organizations that have built their AI workflows on AWS or Google Cloud infrastructure will be able to access OpenAI's models natively, without routing through Azure. For OpenAI, multi-cloud access is a prerequisite for the kind of enterprise scale that justifies its current valuation and funds ongoing compute investments.
For Microsoft, the picture is more nuanced. The company loses the structural advantage that Azure exclusivity provided during the first phase of enterprise AI adoption. It retains a first-mover relationship, a 27% equity stake, and a capped but guaranteed revenue share through 2030. The deal also removes the AGI-related uncertainty that had created a potential liability: under the original terms, an AGI declaration would have required a major renegotiation with unclear outcomes. That risk is now retired.
Both companies described the amendment as a mutual simplification. The question for the broader market is whether the partnership model that helped define enterprise AI's first chapter will continue to hold that position as OpenAI expands its reach across the cloud landscape.
This article is based on reporting from CNBC, Ars Technica, and the official announcements from OpenAI and TechCrunch.
Image courtesy of appshunter.io and Unsplash.
This article was generated with AI assistance and reviewed for accuracy and quality.
