OpenAI Plans to Buy Google's AI Chips, Report Says

OpenAI plans to buy Google's Tensor Processing Units, or TPUs, directly from Google, The Information reported Tuesday, citing a person familiar with the matter. Reuters carried the report on Wednesday.

The report did not give a price, a quantity, a chip generation, or a delivery date. OpenAI declined to comment to Reuters. Google did not immediately respond to a Reuters request for comment.

Buying the chips outright would be a step beyond renting them. Reuters reported in June 2025 that OpenAI had begun renting Google's AI chips through Google Cloud. Last week, OpenAI said it would make its frontier AI models available through Google Cloud.

The Information also pointed to job postings for engineers to optimize OpenAI's most advanced models for TPUs. That is the kind of hiring a company does when the technical work of moving models to new hardware is already under way.

Why a second supplier matters

OpenAI is one of Nvidia's largest customers. Nvidia's graphics processors lead the market for AI chips, and Google's TPUs have gained ground as an alternative as companies look for cheaper options and chips built for AI work. Google began selling TPUs directly to some customers this year, after years of offering them only through its own cloud.

The timing is not an accident. On Tuesday, Oracle announced at its AI World conference that Oracle Cloud will offer Google's TPUs starting in calendar 2027. Oracle called it the first hyperscale cloud outside Google to offer the chips, and an Oracle executive described the deal as a multi-year, multi-gigawatt partnership.

OpenAI is not cutting Nvidia loose. Nvidia agreed in August to provide more than $100 billion in credit support for a planned OpenAI data center project in Ohio, according to reports on the deal. The reported Google purchase sits alongside that commitment, not in place of it.

The spending behind it

Amazon, Microsoft, and Google parent Alphabet plan to spend as much as $600 billion combined this year on AI infrastructure, Reuters reported. Investors have started asking when that spending turns into returns, which is why chip supply has become a board-level question for AI companies.

Alphabet reports quarterly results later on Wednesday. Investors will be watching for signs that its heavy AI spending is still driving growth in its cloud business, and for any word on TPU sales. Alphabet has said it expects to begin recognizing a small share of revenue from TPU sales agreements by the end of this year, with most of it expected in 2027.

For now, the plan is reported, not confirmed. Neither company has commented on the record, and no purchase order has surfaced. If one does, it will mark a shift in how the company behind ChatGPT powers its models: still heavily dependent on Nvidia, but no longer dependent on Nvidia alone.