OpenAI’s secret weapon against Nvidia dependence takes shape
1 min read
Summary
OpenAI is in the final stages of designing its AI chip with the goal of reducing its reliance on Nvidia hardware, according to a Reuters report.
It plans to send its designs to Taiwan Semiconductor Manufacturing Co (TSMC) for production in the next few months, however, the chip is yet to be officially announced.
The exact capabilities, technical details and timeline of the chip are unknown, but OpenAI is thought to want to improve the design over time to give it more leverage in negotiations with chip suppliers and perhaps grant the company future independence with a chip it controls outright.
Major tech companies like Microsoft, Amazon, Google and Meta have created their own AI acceleration chips to reduce costs and mitigate shortages of AI chips supplied by Nvidia, which enjoys a near-monopoly on high-powered GPUs for data centre use.
OpenAI’s customised chip project has been in the works for some time and in early 2024, OpenAI CEO Sam Altman began spending considerable time travelling the world trying to raise up to $7tn to increase chip fabrication capacity.
