Nvidia, look away! OpenAI is almost ready to deliver first prototype of its AI GPU – General Processing Unit
Posted by:
Sarah Collins
Tech giants such as Microsoft, Google, and Meta are working on developing their own artificial intelligence chips to reduce reliance on Nvidia. OpenAI, also part of the $500 billion Stargate initiative, is nearing completion of its first AI chip, which is expected to enter mass production by 2026. The chip, being developed by a small team led by Richard Ho, will initially be used for running AI models and improving negotiating positions with suppliers like Nvidia. Utilizing TSMC’s advanced 3-nanometer process, the chip will feature a systolic array architecture, HBM, and advanced networking capabilities. Talks with Broadcom were held to potentially design and build the chip.
Recent Posts1
