OpenAI Cuts Exclusive Microsoft Cloud Deal, Expands AWS Partnership

April 29, 2026 — According to industry media reports, OpenAI has officially restructured its partnership with Microsoft, ending the long-standing exclusive cloud service agreement. Under the new terms, OpenAI is no longer restricted to offering its models solely through Microsoft Azure and can now freely deploy its AI models on platforms including AWS and Google Cloud. At the same time, Microsoft will no longer pay revenue-sharing fees to OpenAI based on cloud sales; all revenue generated from selling OpenAI models through Azure will be retained entirely by Microsoft.

This structural adjustment marks OpenAI's "independence" from its deep ties to Microsoft and the swift expansion of its collaboration with Amazon Web Services. AWS stated in an official blog: "This marks the beginning of deeper collaboration between AWS and OpenAI." The two companies will jointly continue to roll out cutting-edge models and agent capabilities on Amazon Bedrock, ensuring that customer-built applications benefit from the latest technological breakthroughs. Additionally, OpenAI's Codex tool will also be available on AWS for building production-grade AI agents.

Sources close to the negotiations reveal that Google Cloud is also actively evaluating the new agreement terms and may pursue collaboration with OpenAI. Analysts believe this shift not only significantly broadens OpenAI's enterprise distribution channels, but also puts an end to the previously controversial clause that would change the business relationship upon achieving Artificial General Intelligence (AGI). For Microsoft, losing exclusivity while retaining full profit from cloud-based sales makes the overall deal more balanced.

Brief Analysis: OpenAI's pivot from a single cloud dependency to a multi-cloud strategy signals a push for greater ecosystem openness. The swift joint launch with AWS shows that AI giants are accelerating their move toward open platforms. In the future, enterprise customers are likely to gain more flexible access to top-tier AI capabilities across multiple cloud environments, also pushing cloud vendors into fiercer competition at the model service level.