OpenAI is making a strategic bet that could fundamentally reshape how enterprises access artificial intelligence. In an internal memo viewed by multiple outlets, the company's newly appointed Chief Revenue Officer Denise Dresser outlined how OpenAI plans to reduce its reliance on Microsoft and expand through Amazon Web Services (AWS), signaling that the long-standing partnership with Microsoft has become a limiting factor rather than an asset .

Why Is OpenAI Suddenly Distancing Itself From Microsoft?

For years, Microsoft's $13 billion investment in OpenAI seemed like a perfect marriage. Microsoft got exclusive access to cutting-edge AI models for its products, while OpenAI secured a powerful cloud infrastructure partner. But the relationship has quietly fractured. In her memo, Dresser acknowledged the partnership's historical importance while being blunt about its constraints: "Our Microsoft partnership has been foundational to our success. But it has also limited our ability to meet enterprises where they are, for many that's Bedrock" . Bedrock is AWS's platform that gives companies access to multiple AI models, including OpenAI's, through a single interface.

Dresser

The timing matters. In mid-2024, Microsoft added OpenAI to its official list of competitors in its annual report, a roster that previously included only megacap tech firms like Amazon, Apple, Google, and Meta. That same year, Microsoft began publicly testing its own homegrown AI model, signaling it was no longer content to rely solely on OpenAI for AI capabilities. OpenAI, meanwhile, has been quietly diversifying its cloud partnerships, working with providers like CoreWeave, Google, and Oracle for computing capacity .

What Does the Amazon Partnership Actually Mean for Customers?

Since OpenAI announced its partnership with Amazon in late February 2026, the response has been overwhelming. Dresser noted that "inbound demand from our customers for this offering has been frankly staggering" . This isn't just about availability; it's about flexibility. Enterprises increasingly want to choose their cloud provider rather than being locked into Microsoft's ecosystem. AWS's dominance in cloud infrastructure means that many large organizations already run their operations there, making it natural for them to access AI tools through the same platform.

Dresser

The Amazon Stateful Runtime Environment, a new capability being rolled out, adds another layer of appeal. By enabling memory, context, and continuity across interactions, OpenAI's models can now operate reliably over time and across complex business processes, rather than treating each conversation as isolated. This transforms AI from a stateless tool into something that can genuinely integrate into enterprise workflows .

How Is OpenAI Positioning Itself Against Anthropic in Enterprise?

The real competitive battle isn't with Microsoft anymore; it's with Anthropic, the AI startup founded by former OpenAI researchers. Anthropic's Claude model has become the enterprise favorite, a phenomenon so pronounced that venture capitalists at industry conferences are calling it "Claude mania." Dresser acknowledged this directly in her memo, noting that "the market is as competitive as I have ever seen it" . Both companies are preparing for initial public offerings as soon as 2026, and enterprise revenue is critical to their valuations.

Dresser's memo also took aim at Anthropic's accounting practices, alleging that the company inflates its stated run rate by roughly $8 billion through accounting treatments that gross up revenue from cloud partnerships. OpenAI, by contrast, reports Microsoft revenue share net, which it argues is more aligned with public company standards. Dresser also criticized Anthropic for not acquiring enough computing capacity, claiming the company is "operating on a meaningfully smaller curve" compared to OpenAI .

OpenAI's counter-strategy is to position itself as a platform company rather than a single-product vendor. Dresser emphasized that "multi-product adoption makes us harder to replace" and that the company should "stop thinking like a company with separate product lines" and instead "think like a platform company with multiple entry points and one integrated enterprise offering" . This means bundling models, agents, and enterprise tools into a cohesive system that becomes increasingly difficult for customers to abandon.

Dresser

Steps to Understanding OpenAI's New Enterprise Strategy

• Model Leadership: OpenAI is prioritizing its newest models, including one codenamed Spud, which the company claims delivers stronger reasoning and more reliable output for professional work than competitors.
• Agent Platform Dominance: The company is positioning its Frontier platform as the default system for enterprise agents, aiming to own the orchestration, control, observability, and governance layers that enterprises need.
• Multi-Cloud Distribution: Rather than relying solely on Microsoft, OpenAI is now available through AWS Bedrock, Google Cloud, and Azure, giving enterprises genuine choice in how they access the technology.
• Switching Cost Elevation: By embedding OpenAI's tools deeper into enterprise workflows and offering multi-year, multi-product contracts, the company aims to raise the cost of switching to competitors.

The stakes are enormous. Dresser noted that enterprise AI is entering a more mature phase where raw capability alone is no longer enough. Customers now demand fit, integration, trust, and the ability to deploy and improve systems at scale. OpenAI's biggest constraint right now, according to Dresser, is not demand but capacity, which is why the company is prioritizing talent acquisition in the second quarter of 2026 .

What's happening here is a fundamental realignment of the AI industry. OpenAI is no longer the scrappy startup backed by Microsoft; it's becoming an independent platform company that happens to work with multiple cloud providers. For enterprises, this shift means more choice and less lock-in. For Microsoft, it signals that its exclusive partnership with OpenAI was always temporary. And for Anthropic, it means OpenAI is no longer playing defense but aggressively consolidating its enterprise position before both companies go public.