Microsoft Plans to Sue Amazon and OpenAI: $50 Billion Cloud Partnership Allegedly Breached

Microsoft and OpenAI’s relationship continues to deteriorate, extending into legal disputes.

According to media reports on the 18th, Microsoft is considering seeking legal remedies over a roughly $50 billion partnership between Amazon and OpenAI, with the core dispute centered on whether this deal infringes on Microsoft’s exclusive rights to access OpenAI’s API. Citing insiders, the media said the three parties are still negotiating an out-of-court resolution, but Microsoft has taken a firm stance—“If they breach, we will sue.”

The dispute focuses on OpenAI’s new enterprise product Frontier. This product is the centerpiece of the collaboration announced last month between Amazon and OpenAI, which also committed to purchasing $138 billion worth of cloud services from Amazon Web Services (AWS).

Microsoft believes that regardless of how Amazon and OpenAI structure their technology, bypassing Azure routing API requests is not feasible contractually and violates the spirit of the agreement. This dispute poses a direct threat to OpenAI’s plans to go public this year.

Contract Dispute Core: API Exclusivity Clause

In 2019, Microsoft invested $1 billion in OpenAI and has long served as its exclusive cloud provider. Last October, Microsoft approved OpenAI’s corporate restructuring, relinquishing its overall exclusive cloud status but retaining a key clause: All calls to OpenAI models via application programming interfaces (APIs) must be routed through the Microsoft Azure platform.

Frontier is the trigger for the dispute. The product deploys an AI agent fleet—robots that can operate independently under human commands—to serve enterprise clients. Amazon and OpenAI jointly developed a system called “Stateful Runtime Environment” (SRE), running on Amazon’s Bedrock AI platform. Both companies claim that this system, which accesses enterprise data stored on AWS to give AI agents memory and context capabilities, is “stateful” and does not constitute a direct API call to OpenAI’s “stateless” base models, thus potentially bypassing Microsoft’s exclusivity.

Microsoft does not recognize this. According to reports, Microsoft technical experts believe that within the current contractual framework, running Frontier without going through Azure is technically unfeasible. “We understand our contract,” a person familiar with Microsoft’s position said, “If Amazon and OpenAI want to bet on their contract lawyers’ creativity, I support us, not them.”

Internal Disputes and Deliberate Wording

An internal memo obtained by the media states that Amazon has issued strict guidelines to employees, restricting how they describe the SRE product to avoid provoking Microsoft.

According to the memo, AWS staff can tell customers that SRE is “powered by OpenAI,” “enabled by OpenAI,” or “integrates with OpenAI,” but are explicitly prohibited from using phrases like “enables access” or “calls on” ChatGPT or other models, nor can they imply that the most advanced models on AWS are callable.

Sources familiar with the matter revealed that lawyers from all three parties had engaged in intense negotiations for weeks over the scope of the Amazon agreement and how it is described. When the three companies issued statements about Frontier, Microsoft insisted it remains the exclusive cloud provider for OpenAI API, with no change from last October’s restructuring.

OpenAI maintains that its partnership with Amazon does not create a backdoor for calling its stateless base models, and that it has the right to develop new products with third parties, provided those products do not primarily offer API access. The company also believes that Microsoft is unlikely to take legal action at a sensitive time when regulatory investigations into Azure’s anti-competitive practices are underway in the US, UK, and EU. Microsoft responded, “We believe OpenAI understands and respects the importance of fulfilling its legal obligations.” Both Amazon and OpenAI declined to comment.

IPO Timeline Under Pressure

This legal dispute is particularly unfavorable for OpenAI. The company initially aimed for an IPO this year, but ongoing litigation makes this timeline uncertain.

OpenAI recently completed a funding round valuing it at $110 billion, but it still needs to raise more capital to cover the enormous costs of training and operating large language models. Meanwhile, its IPO process has become more complicated due to Elon Musk’s lawsuit against CEO Sam Altman—Musk and Altman co-founded OpenAI in 2015, and Musk now accuses Altman of abandoning the nonprofit mission for personal gain. The trial is scheduled to begin next month in Oakland.

“OpenAI doesn’t need another lawsuit right now,” said a person familiar with Microsoft’s stance.

This dispute reflects a deeper evolution in the relationship between Microsoft and OpenAI. As OpenAI actively expands its cloud service partnerships and loosens early contractual restrictions, Microsoft’s largest investor increasingly views it as a competitor in the enterprise AI services space. The AI agent business involved in OpenAI’s Frontier product overlaps heavily with the core services Microsoft provides to enterprises via Azure—OpenAI’s products have also been a key driver of record-breaking Azure revenue.

Risk Warning and Disclaimer

Market risks are present; invest cautiously. This article does not constitute personal investment advice and does not consider individual users’ specific investment goals, financial situations, or needs. Users should evaluate whether any opinions, views, or conclusions herein are suitable for their particular circumstances. Invest at your own risk.