OpenAI continues to strengthen its AI infrastructure to support the rapid expansion of agent-based workloads. On Monday, the creator of ChatGPT announced an agreement with Amazon to purchase $38 billion worth of cloud computing resources over the next seven years.
According to OpenAI, it will begin utilizing AWS computing power right away, aiming to have all planned capacity in place by the close of 2026, with options to further increase usage through 2027 and later years.
This agreement comes after OpenAI restructured last week, a move that removed the requirement to obtain Microsoft’s consent before purchasing computing resources from other providers.
The partnership with Amazon fits into OpenAI’s broader strategy to expand its computing capabilities, with plans to invest over $1 trillion in the coming ten years. The company has also revealed new data center projects with Oracle, SoftBank, the United Arab Emirates, among others, and has established partnerships with chip manufacturers such as Nvidia, AMD, and Broadcom.
Some industry experts interpret the growing investments by OpenAI and other major technology companies as a sign that the sector may be entering an AI bubble, where enormous amounts of money are funneled into developing a technology that remains unproven and potentially risky, with uncertain prospects for significant returns.
