OpenAI plans to buy Google TPU chips, The Information reported on Tuesday, in a move that would put ChatGPT’s maker on its rival’s silicon for the first time as an owner rather than a renter. The report, citing a person familiar with the matter, did not disclose financial terms or a timeline for any purchase.

That’s a striking shift for a company that has built its empire on Nvidia hardware. OpenAI is already one of the largest buyers of Nvidia’s graphics processing units, the chips it uses to train AI models and run them for users. Amazon, Microsoft and Google parent Alphabet plan to spend as much as $600 billion combined this year on AI infrastructure, according to Reuters, and OpenAI’s reported purchase sits right in the middle of that spending race.

Neither side confirmed the deal. Google did not immediately respond to a Reuters request for comment outside regular business hours, while OpenAI declined to comment, Reuters reported. The lack of a denial, plus a public hiring trail, is what gives this report weight beyond a rumor.

OpenAI Google TPU chips: the job posting that backs up the report

The clearest public signal is a job listing. OpenAI is hiring a Software Engineer, TPU Compilers, in San Francisco, with the role described as hiring someone to “lead the optimization” of its frontier models on TPUs. The Information reported that the listing suggests the work is still in its early stages, with experience in either training or inference welcome.

Compiler work matters here. A chip is only as useful as the software that turns a model’s code into instructions it can run fast. Hiring someone to lead that effort means OpenAI isn’t just shopping for hardware. It’s building the team that would make Google’s chips actually work well with its models.

That team already has a Google connection at the top. Richard Ho, who leads OpenAI’s hardware division, was previously senior director of engineering at Google, where he led development of its Cloud TPUs. OpenAI is also developing its own custom chip, codenamed Jalapeno, with Broadcom, which it plans to begin using later this year.

From renting TPUs to owning them

Buying chips outright would be a step change from how OpenAI has used Google hardware so far. Reuters reported in June 2025 that OpenAI had begun renting Google’s AI chips through Google Cloud to run ChatGPT and other products. That was its first meaningful use of non-Nvidia chips, and the goal at the time was cutting inference costs, the everyday computing used to answer user prompts.

Google, for its part, was not renting its most powerful TPUs to OpenAI at that point. A purchase would change that dynamic. It would give OpenAI direct control over a slice of its compute capacity instead of relying on rented capacity that can disappear when demand spikes.

Google has also changed how it sells TPUs. The company built tensor processing units more than a decade ago for its own workloads, starting with search, and for years kept them largely inside Google Cloud. Google began selling TPUs directly to some customers this year, turning an internal tool into a product line.

Why this matters

Rivals in AI are becoming each other’s suppliers, whether they like it or not. Last week, OpenAI said it would make its frontier AI models available through Google Cloud, adding a distribution deal to what could now become a hardware deal. OpenAI and Google compete directly in AI assistants and search, yet they keep signing commercial agreements with each other.

Anthropic and Oracle are moving the same way. Anthropic already taps Google TPUs, and Oracle said on Tuesday that it will offer Google TPUs through its cloud, making Google’s chips available outside Google’s own infrastructure in a new way. For Google, each of those deals turns a cost center, its chip program, into a revenue line.

Alphabet reports quarterly results later on Wednesday, October 28, and investors will be watching for signs that heavy AI spending is driving growth in its cloud business. Chief Financial Officer Anat Ashkenazi has previously said Alphabet expects to begin recognizing a small percentage of revenue from TPU sales agreements by the end of this year, with the vast majority expected in 2027. An OpenAI purchase, if confirmed, would land squarely in that story.

For Nvidia, the risk isn’t losing OpenAI tomorrow. It’s losing exclusivity. OpenAI will keep buying Nvidia GPUs for training, where Nvidia remains dominant. But inference, where models serve hundreds of millions of users, is where costs pile up and where rivals can win on price or efficiency for specific jobs. If OpenAI proves its models run well on TPUs at scale, other labs and cloud customers will have a tested alternative to point to.

What to watch next

Three things will tell us how serious this is. First, whether OpenAI or Google confirms a purchase, and at what scale. Second, whether more TPU compiler and hardware roles appear, which would signal a larger team than a single hire. Third, what Alphabet says about TPU revenue on Wednesday’s earnings call, and whether it names new external customers.

Until then, treat this as a reported plan, not a done deal. The job posting is real, the rental history is real, and the commercial ties between OpenAI and Google keep getting denser. A purchase would be the next logical step, but in AI infrastructure, logical steps still need contracts, delivery dates, and software that works.

FAQ

Is OpenAI buying Google TPU chips?

The Information reported on October 27, 2026, that OpenAI plans to buy Google’s TPU chips, citing a person familiar with the matter. Neither company has confirmed the plan, and no financial terms have been disclosed.

Why would OpenAI buy Google TPU chips instead of Nvidia GPUs?

OpenAI is already one of Nvidia’s biggest customers, but TPUs give it a second hardware option, especially for inference work where answering user prompts at scale drives much of its computing cost.

Has OpenAI used Google TPUs before?

Yes. Reuters reported in June 2025 that OpenAI had begun renting Google TPUs through Google Cloud to run ChatGPT and other products, its first meaningful use of non-Nvidia chips.

Does this mean OpenAI is dropping Nvidia?

No. OpenAI still buys large volumes of Nvidia GPUs and is also developing its own chip with Broadcom. Buying Google TPUs would add another supplier, not replace Nvidia overnight.

Sources: Reuters, The Information, Bloomberg