Amazon’s $38 Billion Deal With OpenAI
Amazon stunned the tech world with a $38 billion deal for OpenAI, teaming up with the company to supercharge cloud computing and generative AI and establishing itself as a leading force in artificial intelligence.
Amazon has made a significant return to the AI arena. After being criticized for lagging in the artificial intelligence race, the e-commerce and cloud giant recently signed a massive multi-year deal worth $38 billion with OpenAI, the company behind ChatGPT. The partnership marks a major turning point for Amazon Web Services (AWS), positioning it as a central player in powering the next generation of AI technology.
Under the agreement, OpenAI will run its advanced AI workloads—including large language models and future versions of ChatGPT—on AWS’s powerful infrastructure. This means OpenAI will have immediate and growing access to Amazon’s computing resources, including hundreds of thousands of NVIDIA GPUs and the ability to scale to millions of CPUs. The $38 billion commitment will be spread over seven years, allowing OpenAI to massively expand its computing capacity while leveraging the scale, performance, and security of AWS.
For Amazon, this isn’t just another business deal—it’s a comeback story. After dominating the cloud computing industry for years, AWS has recently seen its lead diminish as Microsoft and Google have secured large AI contracts. According to Synergy Research Group, Amazon’s share of the cloud market is expected to fall from 34% to 29% in 2022 after the launch of ChatGPT. Investors began calling Amazon a “laggard” in the AI race, noting that it had not released a major large language model or a consumer chatbot like ChatGPT or Google Gemini. But that perception is now changing rapidly.
The new deal makes OpenAI one of AWS’s most high-profile customers. In practice, this gives OpenAI access to Amazon EC2 UltraServers, which are specifically designed for heavy AI workloads. These servers combine massive clusters of NVIDIA’s GB200 and GB300 chips, providing lightning-fast performance for training and deploying AI models. This setup will allow OpenAI to run everything from inference tasks to training an entirely new generation of models—helping ChatGPT respond to users in real time.
“Scaling frontier AI requires massive, reliable computation,” said Sam Altman, CEO of OpenAI. “Our partnership with AWS strengthens the broader computing ecosystem that will power this next era and bring advanced AI to everyone.”
AWS CEO Matt Garman echoed this sentiment: “As OpenAI continues to push the boundaries of what is possible, AWS’s best-in-class infrastructure will serve as the backbone for their AI ambitions.”
This partnership builds on the companies’ previous collaborations. Earlier this year, OpenAI’s Foundation models became available through Amazon Bedrock—AWS’s AI platform, which lets developers build and optimize applications using pre-trained models. OpenAI’s models quickly became one of the most popular choices, with thousands of AWS customers, including Peloton, Thomson Reuters, and comScore, integrating them into their products and services for coding, data analysis, and workflow automation.
This $38 billion deal also signals that OpenAI is no longer locked exclusively to Microsoft’s Azure cloud. From 2019 to 2023, OpenAI had a special dispensation that required it to use only Microsoft’s infrastructure unless otherwise permitted. But faced with limitations in access to sufficient computing resources, OpenAI negotiated new flexibilities. Just last week, Microsoft and OpenAI restructured their agreement, freeing OpenAI to buy cloud services from any provider—and Amazon was ready to seize the opportunity.
The timing couldn’t be better for Amazon. The company has been investing heavily in AI infrastructure, recently unveiling “Project Rainier,” an $11 billion AI data center in Indiana where AI startup Anthropic trains models using Amazon’s custom Trainium chips. This new partnership with OpenAI solidifies AWS’s growing momentum after months of stagnation and criticism. Investors immediately took notice—Amazon’s stock jumped 5% to a record high after the announcement.
Analysts see this as an important step in Amazon’s effort to get stronger again. Mamta Valechha, an analyst at Quilter Cheviot, said, “While it is small relative to other deals OpenAI has made with other cloud providers, it represents a key first step in Amazon’s effort to partner with a company that is spending over a trillion dollars on computing power in the coming years.”
Indeed, the numbers are staggering. Microsoft recently revealed a $250 billion commitment to OpenAI for Azure cloud services, while Oracle signed a $300 billion deal with the same startup. Meanwhile, Google has struck multi-billion-dollar chip and cloud partnerships with Anthropic and other AI firms. With this $38 billion deal, Amazon is showing that it is ready to compete head-to-head with the biggest players in the AI arms race.
Still, the road to AI dominance has not been easy for Amazon. The company has dealt with a number of executive departures, including a key vice president overseeing generic AI projects, who left earlier this year. To tighten operations and cut bureaucracy, CEO Andy Jassy has streamlined management layers and even introduced an anonymous complaint system to flag inefficiencies. At the same time, Amazon has made its largest-ever reduction in its workforce, laying off nearly 14,000 corporate employees. These moves are part of a broader strategy to focus spending where it matters most—on AI infrastructure and innovation.
TThat focus comes with a hefty price tag. Amazon’s capital spending for 2025 is projected to reach approximately $125 billion, significantly higher than in previous years and roughly equal to Microsoft’s expected spending. That’s even more than Alphabet’s estimated $93 billion. But analysts believe the OpenAI deal could help Amazon recoup a large portion of that investment. Brian Pietz, an analyst at BMO Capital Markets, estimates that AWS’s backlog could grow by about 20% in the fourth quarter of 2025, from $200 billion to about $240 billion.
“It clearly seems like they (Amazon) are finally more in the flow of what’s happening with these larger language models than before,” said William Lee, an investor at Suro Capital, which has equity in OpenAI.
Beyond the financial, the partnership highlights broader changes in the tech landscape. As generative AI is reshaping industries, the companies providing the underlying computing power, such as Amazon, Microsoft, Google, and Oracle, have become as important as the AI developers themselves. AWS’s infrastructure, which supports clusters of more than half a million chips, is one of the few capable of handling the enormous scale required for AI training and deployment. This gives Amazon a significant edge as demand for compute continues to grow.
For OpenAI, the deal represents another step toward diversification and independence. By partnering with multiple cloud providers, OpenAI can avoid disruptions and improve resiliency, ensuring its AI models remain online and accessible even under heavy global demand. For Amazon, it’s a chance to establish dominance in the market it once led—and prove to Wall Street that it’s not only playing catch-up but capable of leading the next era of AI development.
Ultimately, this isn’t just a deal between two tech giants—it’s emblematic of how the AI gold rush is reshaping alliances and fortunes in Silicon Valley. OpenAI has secured more computing power to power its cutting-edge models, while Amazon has secured its place in the conversation as a top-tier AI powerhouse.
After months of doubt, Amazon has made one thing clear with its $38 billion move: it’s no longer lagging behind in AI—it’s back in the race, and it’s betting big on the future of artificial intelligence.