OpenAI is entering the final stages of designing its long-rumored AI processor with the aim of decreasing the company’s dependence on Nvidia hardware, according to a Reuters report released Monday. The ChatGPT creator plans to send its chip designs to Taiwan Semiconductor Manufacturing Co. (TSMC) for fabrication within the next few months, but the chip has not yet been formally announced.
The OpenAI chip’s full capabilities, technical details, and exact timeline are still unknown, but the company reportedly intends to iterate on the design and improve it over time, giving it leverage in negotiations with chip suppliers—and potentially granting the company future independence with a chip design it controls outright.
In the past, we’ve seen other tech companies, such as Microsoft, Amazon, Google, and Meta, create their own AI acceleration chips for reasons that range from cost reduction to relieving shortages of AI chips supplied by Nvidia, which enjoys a near-market monopoly on high-powered GPUs (such as the Blackwell series) for data center use.
In October 2023, we covered a report about OpenAI’s intention to create its own AI accelerator chips for similar reasons, so OpenAI’s custom chip project has been in the works for some time. In early 2024, OpenAI CEO Sam Altman also began spending considerable time traveling around the world trying to raise up to a reported $7 trillion to increase world chip fabrication capacity.
A large investment
The path to creating a custom AI chip requires substantial resources. Industry experts told Reuters that designing a single version of such a processor could cost as much as $500 million, with additional expenses for developing supporting software and hardware potentially doubling that amount.
The current OpenAI chip project, led by former Google chip designer Richard Ho, involves a team of 40 engineers working with Broadcom on the processor design, according to Reuters. The Taiwanese company TSMC, which also produces Nvidia’s chips, will manufacture OpenAI’s chips using its 3-nanometer process technology. The chips will reportedly incorporate high-bandwidth memory and networking features similar to those found in Nvidia’s processors.
Initially, OpenAI’s first chip will focus primarily on running AI models (often called “inference”) rather than training them, with limited deployment across the company. The timeline suggests mass production could begin at TSMC in 2026, though the first tape-out and manufacturing run faces technical risks that could require additional fixes and could delay the project for months.
OpenAI’s move into AI hardware comes as major tech companies spend record amounts on AI infrastructure. Microsoft plans to invest $80 billion in 2025, while Meta set aside $60 billion for the next year, Reuters notes. Last month, OpenAI (working with SoftBank, Oracle, and MGX) announced a new $500 billion “Stargate” infrastructure project aimed at building new AI data centers in the US.