Microsoft and OpenAI Rewrite Their Deal: What It Means for Enterprise AI

The Microsoft-OpenAI partnership has been one of the most watched, and at times strained, relationships in enterprise tech. This week, the two companies announced a restructured agreement that effectively ends OpenAI’s exclusivity on Microsoft’s Azure cloud, while keeping the financial ties intact. For business leaders evaluating AI infrastructure, the changes signal a new era of flexibility—and competition.

Under the revised terms, OpenAI can now offer its models—including Codex and future releases—through any cloud provider, including AWS, Microsoft’s biggest rival. Amazon had already inked a $50 billion deal with OpenAI earlier this year, which reportedly rankled Microsoft. The new arrangement removes that exclusivity roadblock, allowing OpenAI to meet enterprise customers where they are, whether on Azure, AWS, or potentially Google Cloud.

What does Microsoft get in return? A steady revenue stream: Microsoft retains a 20 percent cut of OpenAI’s revenue from ChatGPT and its API platform, including sales made through competing clouds. Additionally, Microsoft’s non-exclusive license for OpenAI models now extends through 2032, and the company no longer has to pay a 20 percent share of its Azure OpenAI revenue back to OpenAI. Microsoft also holds roughly 27 percent of OpenAI’s for-profit arm.

Perhaps the most significant change is the removal of the AGI clause. Previously, if OpenAI ever achieved artificial general intelligence, Microsoft would lose access to its most advanced models. That clause is gone, meaning Microsoft can continue using future OpenAI breakthroughs regardless of AGI milestones. For enterprises, this reduces uncertainty around model availability and long-term AI strategy.

Behind the scenes, tensions had been building for years—from Sam Altman’s brief ouster in 2023 to disputes over OpenAI’s for-profit restructuring and acquisitions like Windsurf. Microsoft’s hiring of DeepMind cofounder Mustafa Suleyman to lead its own AI efforts further strained the relationship. The new deal effectively distances Microsoft from OpenAI’s internal drama while keeping the financial upside.

For business leaders, the takeaway is clear: The era of single-cloud AI lock-in is fading. Microsoft is hedging its bets, investing in internal models like MAI-Transcribe-1 and even trialing Anthropic’s Claude and Google’s Gemini for certain tasks. The partnership with OpenAI is now more about revenue sharing than exclusive collaboration. As enterprises build their AI stacks, they should expect a multi-model, multi-cloud reality—where the best tool for the job, not the vendor relationship, drives decisions.

Source: The Verge