OpenAI set to start mass production of its own AI chips with Broadcom

OpenAI is preparing to launch its first artificial intelligence chip next year through a partnership with U.S. semiconductor giant Broadcom, according to a report in the Financial Times citing people familiar with the project.

Rather than selling the chip to outside customers, OpenAI intends to use it internally to power its own models and services, the report said. The company, which has helped bring generative AI into the mainstream with products like ChatGPT, relies on massive computing power to train and operate its systems.

The effort is part of OpenAI’s broader push to reduce its reliance on Nvidia, the current dominant supplier of AI chips. Last year, Reuters reported that OpenAI had been working with both Broadcom and Taiwan Semiconductor Manufacturing Co. (TSMC) to design its own silicon, while also blending AMD chips into its infrastructure to meet growing demand. Sources at the time said OpenAI aimed to finalize the design for its first-generation chip and send it to TSMC for fabrication.

Broadcom, meanwhile, sees AI as a major driver of its future business. CEO Hock Tan said Thursday that the company expects revenue from AI-related products to grow “significantly” in fiscal 2026 after securing more than $10 billion in infrastructure orders from a new customer he did not name.