OpenAI and Broadcom have been collaborating for 18 months on a new line of co-designed chips optimized for inference and networked through Broadcom’s Ethernet stack. The companies announced plans to develop and deploy racks of OpenAI-designed chips starting late next year.
Breakdown
- OpenAI and Broadcom have agreed to deploy 10 gigawatts of custom AI chips over four years.
- The deal involves co-designing custom accelerators, with no equity exchanged between the companies. 34s
- The new chips are expected to be about 30 percent cheaper than current GPU options. 49s
- OpenAI aims to control the full hardware stack to improve efficiency and performance. 1m 4s
- This agreement is part of OpenAI's broader strategy to secure infrastructure and stay ahead of competitors. 1m 10s
See The Full Story

Rise of AI
The release of Open AI's ChatGPT has accelerated the development of artificial intelligence and warnings about its impact on society.