OpenAI, the developer behind ChatGPT, has signed a landmark $38 billion agreement with Amazon Web Services (AWS), further expanding its global cloud infrastructure network as part of a sweeping series of high-value partnerships in 2025.
The seven-year deal will provide OpenAI with access to extensive computing resources, including hundreds of thousands of Nvidia GPUs — the core technology driving the generative AI revolution — along with tens of millions of conventional CPUs for large-scale deployment of “agentic AI.”
Under the agreement, OpenAI will begin utilizing AWS capacity immediately, with all computing resources expected to be in full use by the end of 2026. The pact also allows for future expansion as the company scales its next generation of AI models.
“Scaling frontier AI requires massive, reliable compute,” said OpenAI co-founder and CEO Sam Altman in a joint statement. “Our partnership with AWS strengthens the broad compute ecosystem that will power this next era and bring advanced AI to everyone.”
The AWS partnership builds on earlier collaboration between the two companies, with several of OpenAI’s open-source models already hosted on Amazon servers. It also marks OpenAI’s first major deal since restructuring its organization to allow greater commercial flexibility and profit generation, moving beyond its original non-profit framework.
Industry analysts estimate that OpenAI has now committed close to $1 trillion in infrastructure partnerships in 2025 alone, including a $300 billion deal with Oracle and a $500 billion “Stargate” project with Oracle and Japan’s SoftBank.
Despite expectations that OpenAI’s revenues will reach tens of billions of dollars this year — an impressive figure for a company of its age — the scale of its infrastructure spending underscores the enormous costs of powering its advanced AI models.
The latest agreement also deepens the rivalry between Microsoft, which owns a significant stake in OpenAI, and Amazon, its cloud competitor. Yet, for OpenAI, the deal ensures access to one of the world’s most robust cloud ecosystems — a critical step in maintaining its lead in the generative AI race.
