OpenAI has finalized a seven-year, $38 billion partnership with Amazon Web Services (AWS) to expand the computing infrastructure needed for its advanced artificial-intelligence models.
The move represents one of the largest AI-focused cloud agreements ever signed.
Announced on Monday, November 3, 2025, the deal gives OpenAI access to hundreds of thousands of AI-optimized processors, including Nvidia GPUs, across AWS’s global data-centre network.
Work is set to begin immediately, with full operational capacity projected by December 31, 2026, and additional scaling planned for 2027.
The agreement follows OpenAI’s internal restructuring earlier in 2025, which ended its exclusive reliance on Microsoft’s Azure cloud services.
This shift allows the company to diversify its infrastructure and collaborate with multiple cloud providers.
AWS described the partnership as a demonstration of its ability to support “frontier-scale” AI workloads.
READ ALSO: Meta Strikes $10 Billion Cloud Deal With Google to Power AI Push
OpenAI Chief Executive Officer Sam Altman said:
“Building next-generation AI requires massive, dependable compute. Our collaboration with AWS strengthens the ecosystem that will enable this new era and make advanced AI more widely accessible.”
Experts say the deal not only strengthens OpenAI’s ability to train and deploy cutting-edge AI models but also reinforces AWS’s position as a top provider of AI infrastructure, intensifying competition with Microsoft and Google Cloud.
Observers also note that commitments of this scale bring logistical, financial, and environmental challenges, given the enormous energy demands of high-performance data centres.