TL;DR
OpenAI signed a $38 billion, multi-year deal with AWS to run and scale its AI workloads using Amazon’s high-performance cloud infrastructure.
The partnership gives OpenAI access to hundreds of thousands of NVIDIA GPUs and the ability to scale up to tens of millions of CPUs by 2026.
It marks a major shift from OpenAI’s reliance on Microsoft Azure, positioning AWS as a key player in large-scale AI infrastructure.
OpenAI has signed a $38 billion, multi-year deal with Amazon Web Services (AWS) that marks one of the largest technology partnerships in recent years. The agreement will see OpenAI run its advanced artificial intelligence workloads on AWS infrastructure, granting it access to massive computing power built on thousands of the latest NVIDIA chips and scalable cloud clusters.
This collaboration signals a notable shift in how OpenAI manages its operations. For years, it relied heavily on Microsoft’s Azure cloud. Now, by bringing AWS into its compute strategy, OpenAI is diversifying its infrastructure and strengthening its capacity to meet the growing demand for generative AI models like ChatGPT.
AWS says it will provide OpenAI with Amazon EC2 UltraServers featuring hundreds of thousands of chips, with the option to expand to tens of millions of CPUs over time. The deployment will begin immediately and is expected to be completed by the end of 2026, with room for additional growth through 2027. The partnership will allow OpenAI to scale its next-generation model training and inference workloads more efficiently while benefiting from AWS’s reputation for reliable performance, scalability, and security.

Expanding compute access for the AI era
The rapid rise of generative AI has placed enormous pressure on computing resources. Training and operating large models require substantial energy and hardware capacity, and access to GPUs has become a competitive advantage in itself. OpenAI’s move to integrate AWS as a core infrastructure provider is a direct response to that growing demand.
AWS has long experience running large-scale AI infrastructure securely, with clusters that exceed half a million chips. By integrating NVIDIA’s latest GB200 and GB300 GPUs into its EC2 UltraServers, AWS can deliver low-latency performance and more efficient interconnection between systems. That means faster training times and smoother delivery for services like ChatGPT.
Also Read: Applications open for the 2026 Keeling Curve Prize with $500,000 to support global climate solutions
Sam Altman, OpenAI’s co-founder and CEO, said, “Scaling frontier AI requires massive, reliable compute. Our partnership with AWS strengthens the broad compute ecosystem that will power this next era and bring advanced AI to everyone.”
Matt Garman, CEO of AWS, added that OpenAI’s continued expansion “pushes the boundaries of what’s possible,” and that AWS’s infrastructure will act as a backbone for those ambitions.
The deal also builds on earlier collaboration between the two companies. OpenAI’s open-weight foundation models are already available through Amazon Bedrock, allowing AWS customers to access and deploy OpenAI technology in their own systems. According to AWS, thousands of organizations are already using these models for coding, analytics, and scientific applications.





Leave a Reply