OpenAI to develop Its own AI accelerators

Tech giant OpenAI has announced a partnership with Broadcom to begin producing its own hardware systems designed to accelerate the performance of artificial intelligence models.

Through this collaboration, OpenAI aims to build the infrastructure needed to scale AI capabilities and integrate them more efficiently into real-world products and business applications. Developing its own chips will also help the company reduce dependence on external suppliers and increase computational efficiency.

The two companies have been long-time partners, and this new phase of cooperation will focus on building advanced accelerators and networking systems for the next generation of AI formations.

OpenAI will lead the design and architecture of the chips, while Broadcom will oversee technology integration and large-scale manufacturing. The resulting systems are expected to deliver up to 10 gigawatts of computing capacity. The project is set to begin in the second half of 2026 and reach full operational deployment by the end of 2029.