Inside Amazon’s $38 Billion Deal with OpenAI for Nvidia Chips: A New Front in the AI Arms Race

Amazon has struck a $38 billion deal with OpenAI to supply computing power for the ChatGPT-maker’s expanding artificial intelligence operations — a landmark partnership that underscores Amazon’s renewed push into the AI infrastructure race.

Under the seven-year agreement, OpenAI will rely on Amazon Web Services (AWS) to access hundreds of thousands of Nvidia GPUs, including the latest GB200 and GB300 AI accelerators. The move positions AWS as a key player in powering OpenAI’s next generation of large language models and advanced AI tools.

🚀 A Major Win for Amazon in the Cloud AI Battle

For Amazon, this deal is more than a business agreement — it’s a strong vote of confidence in its cloud computing dominance. AWS, already the world’s largest provider of rented computing power, had previously stayed on the sidelines while Microsoft Azure, Google Cloud, and Oracle signed multi-billion-dollar partnerships with OpenAI.

“As OpenAI continues to push the boundaries of what's possible, AWS’s best-in-class infrastructure will serve as a backbone for their AI ambitions,” said Matt Garman, CEO of Amazon Web Services, in a statement to Bloomberg.

Amazon shares rose 4.5% to $255.29 in early New York trading following the announcement, while Nvidia climbed 3.3% to $209.20, reflecting investor optimism about the long-term AI infrastructure boom.

💰 OpenAI’s Massive Infrastructure Spending

The deal is part of OpenAI’s broader $1.4 trillion infrastructure investment plan, covering data centers, chips, and AI research capacity. The sheer scale of spending has raised concerns about an emerging AI investment bubble, but OpenAI remains determined to expand its global compute capabilities.

According to Bloomberg Intelligence analysts Anurag Rana and Andrew Girard, partnering with AWS could relieve some of OpenAI’s capacity pressure, especially as it diversifies beyond smaller “neo-cloud” providers like CoreWeave, with which OpenAI signed a $22.4 billion deal earlier this year.

🤝 OpenAI’s Growing Cloud Partnerships

Before this agreement, Microsoft — OpenAI’s largest investor — had secured an exclusive cloud partnership worth $250 billion through its Azure platform. Oracle also landed a $300 billion deal, while Google Cloud remains an active infrastructure partner for ChatGPT.

By adding AWS to its network, OpenAI now gains access to one of the world’s largest and most reliable global compute infrastructures, with the flexibility to expand further as demand for AI services soars.

Under the deal, OpenAI will start using AWS immediately, with full capacity deployment expected by end-2026.

“Scaling frontier AI requires massive, reliable compute,” said Sam Altman, CEO of OpenAI. “Our partnership with AWS strengthens the broad compute ecosystem that will power this next era and bring advanced AI to everyone.”

🧠 Amazon’s Broader AI Ambitions

Beyond OpenAI, Amazon continues to deepen its footprint in the AI sector. The company is a major investor in Anthropic, an AI firm founded by ex-OpenAI employees. Last week, Amazon announced that Anthropic’s new data center — powered by its Trainium2 AI chips — had become operational.

Meanwhile, Google is also intensifying competition, revealing plans to supply up to one million custom AI chips to Anthropic under a deal valued in the tens of billions of dollars.

🔍 The Bigger Picture

Amazon’s $38 billion OpenAI deal signals that the AI infrastructure race is accelerating — and the cloud giants are now battling for dominance in a trillion-dollar market. For Amazon, it’s both a comeback and a challenge: to prove AWS remains the most scalable, reliable, and cost-efficient foundation for the next generation of artificial intelligence.