Nvidia's LLM hyperscaler partners are actively developing custom, non-Nvidia chips. Nvidia has a deep interest in open source models to counter the LLM hypersc

Nvidia's LLM hyperscaler partners are actively developing custom, non-Nvidia chips. Nvidia has a deep interest in open source models to counter the LLM hyperscalers. Nvidia also has a big bet on robotics and local AI, which requires model efficiency. I'd worry far more about Dario and Sam regulatory capturing the US market via a "model safety" scare.