Amazon and OpenAI Forge Exclusive Partnership
Amazon and OpenAI have announced a landmark strategic partnership, making AWS the exclusive third-party cloud provider for OpenAI's "Frontier" platform. The deal positions AWS as the premier infrastructure partner for next-gen AI and gives OpenAI deep access to Amazon's global cloud ecosystem.
This partnership is backed by a massive $50 billion investment from Amazon into OpenAI, part of a larger $110 billion funding round that values the AI company at $730 billion before the new capital. Amazon's initial investment will be $15 billion, with an additional $35 billion to follow as certain conditions are met. The deal marks a major strategic shift for OpenAI, which previously had an exclusive cloud computing agreement with Microsoft Azure from 2019 until 2025. That exclusivity became a significant constraint, notably when a GPU shortage staggered the rollout of GPT-4.5, prompting a renegotiation of their contract to allow for a multi-cloud approach. OpenAI's "Frontier" platform, launched on February 5, 2026, is central to the deal. It's an enterprise-grade system for building and managing AI agents, treating them like employees with onboarding processes, specific permissions, and access to shared business data from systems like CRMs and data warehouses. The platform is already in use by major corporations, with HP, Intuit, Oracle, State Farm, Thermo Fisher, and Uber named as confirmed early adopters. Frontier is designed as an open platform, capable of managing AI agents built with models from competitors like Anthropic and Google, not just OpenAI. Beyond cloud hosting, the two companies will co-create a "Stateful Runtime Environment" on Amazon Bedrock. This technology will allow AI agents to maintain context, remember previous interactions, and work across different software tools, addressing a key limitation in current enterprise AI. While Microsoft remains a primary partner with a $135 billion stake, this move solidifies OpenAI's multi-cloud strategy, which now includes deals with Google Cloud and Oracle. The AWS partnership specifically calls for OpenAI to use 2 gigawatts of capacity on Amazon's custom Trainium AI chips. Despite the new AWS partnership, Microsoft's role remains critical. According to Microsoft, their existing agreements ensure that all stateless API calls to OpenAI models from third parties, including Amazon, will continue to be hosted on Azure. Microsoft also retains its exclusive license and access to the intellectual property across OpenAI's models and products.
