OpenAI is reportedly working with Broadcom and TSMC to manufacture its first in-house AI chip focusing on inference tasks by 2026, aiming to diversify its supply of specialised processing power.
The increasing demand for AI infrastructure has caused shortages of Nvidia chips, leading OpenAI to explore alternative chip suppliers.
OpenAI plans to use AMD AI chips alongside Nvidia chips via Microsoft Azure cloud infrastructure deployments.
OpenAI has partnered with Broadcom and secured manufacturing capacity with TSMC to produce its custom-designed chip in 2026.