Amazon Web Services (AWS) announced on Tuesday that it will begin selling OpenAI’s models to its cloud customers. This move follows Microsoft’s agreement to terminate its three-year exclusive reselling arrangement, which had granted Azure sole access to OpenAI’s technology during the initial phase of the generative AI era. "It’s something that our customers have asked for, for a really long time," Matt Garman, AWS’s chief executive, told Bloomberg Television.
Select OpenAI models are now available for preview on AWS, with the more powerful GPT models expected to arrive in the coming weeks. This announcement completes a significant restructuring of one of the AI industry's most impactful partnerships, which commenced in February when Amazon committed up to $50 billion as part of OpenAI’s $110 billion funding round, valuing the ChatGPT maker at $852 billion and marking Amazon’s largest-ever investment in any company. Concurrently, OpenAI pledged to spend $100 billion on AWS computing power and Trainium chips over eight years, consuming two gigawatts of capacity. Notably, the companies also jointly developed a Stateful Runtime Environment for agentic AI on Bedrock.
The core question this deal addresses is not the quality of OpenAI’s models for sale on rival clouds, but whether OpenAI can sell enough of them, across all platforms, to justify its massive promised expenditures. This concern is amplified by a Wall Street Journal report indicating OpenAI missed revenue and user targets, projecting $25 billion in cash burn against $30 billion in revenue, coupled with hundreds of billions in infrastructure commitments to AWS, Azure, and Oracle – commitments that presuppose a growth rate OpenAI has yet to demonstrate.
The restructuring formalized on Monday includes an amendment to the Microsoft-OpenAI partnership agreement, dismantling the exclusive arrangement that characterized the initial AI boom. Microsoft’s license to OpenAI’s intellectual property for models and products, previously exclusive, now becomes non-exclusive but extends through 2032. Microsoft will retain its 20% share of OpenAI’s revenue through 2030, though now subject to a cap. OpenAI is still required to ship its models on Azure first, unless Microsoft is unable or chooses not to support the necessary capabilities. This clause preserves Azure’s early-access advantage while removing the lock-in that previously prevented AWS and Google Cloud customers from accessing OpenAI’s technology via their existing cloud providers. Furthermore, the restructuring eliminates the "AGI clause," an unusual legal provision that would have terminated Microsoft’s commercial rights if OpenAI’s board determined the company had achieved artificial general intelligence. This clause had evolved into a governance liability rather than a practical safeguard, and its removal signals a mutual acceptance of the revised relationship dynamic.