OpenAI is preparing to buy AI chips from Google, according to a person familiar with the matter who spoke with The Information. The step would loosen OpenAI's dependence on Nvidia, the company that supplies most of the hardware behind the AI boom and counts OpenAI among its biggest customers.

Neither company has confirmed the plan, and the price, the number of chips, and the delivery dates have not been disclosed.

The processors OpenAI is reportedly after are Google's tensor processing units, or TPUs—chips designed for one job: the math that AI models run on. Google revealed the chip in 2016, saying it had already run TPUs in its data centers for more than a year.

Reports in June 2025 said OpenAI had begun renting Google's TPUs, an arrangement the company described at the time as early testing with no plans to deploy them at scale. A purchase would mean owning the hardware outright instead of paying Google by the hour—a distinction that carries weight for a company whose chip deployments are already being sized in gigawatts. Other details of the plan have not been made public.

The report gained weight from an OpenAI job listing in San Francisco seeking an engineer to lead the optimization of the company's frontier models on TPUs. The work involves compilers—the software that translates an AI model into instructions a specific chip can run—which is why switching chips is harder than swapping parts. Richard Ho, who leads OpenAI's hardware program, spent nearly nine years at Google and was one of the leaders of its Cloud TPU project before joining OpenAI.

The reported purchase plan is the latest move in a wider reshaping of AI chip alliances: OpenAI would be adding Google to a lineup of chip partners that already includes Nvidia and AMD, even as it develops silicon of its own.

In September 2025, Nvidia signed a letter of intent to invest up to $100 billion in OpenAI as it deployed at least 10 gigawatts of Nvidia systems. A gigawatt—a billion watts—is the unit data center builders now use to size their projects. Two weeks later, AMD gave OpenAI the right to buy up to 160 million AMD shares, roughly 10% of the company, a right tied to OpenAI deploying 6 gigawatts of AMD chips.

Nvidia ultimately put $30 billion into OpenAI's funding round, well below the $100 billion outlined in its letter of intent. CEO Jensen Huang said in March that it might be the last such investment before OpenAI goes public.

OpenAI is also preparing its own hardware. In June, the company unveiled Jalapeño, its first custom chip, built with Broadcom. Designed for inference—the step where a trained model generates answers—the chip is slated for initial deployment by the end of 2026.

Google, for its part, unveiled two eighth-generation TPUs in April: the 8t for training AI models and the 8i for running them. Its performance figures are measured against Google's previous-generation chip, Ironwood, not against Nvidia's hardware.

Anthropic, OpenAI's main rival, signed an agreement with Google and Broadcom in April for multiple gigawatts of next-generation TPU capacity that it expects to come online starting in 2027.

Alphabet according to its CFO, began delivering TPU systems to customers' own data centers in the second quarter and expects most revenue from TPU system sales to arrive in 2027. Alphabet is scheduled to report third-quarter earnings after the U.S. market closes on Wednesday.

Google Cloud also began offering OpenAI's GPT-6 models to select customers in private preview last week, meaning the reported chip purchase would give the two companies ties in both hardware and software.