All ai hardware coverage from The Sideline.
OpenAI has unveiled Jalapeño, its first custom inference chip built with Broadcom—a direct move to reduce its dependence on Nvidia and cut the ballooning cost of running AI models at scale.
Nvidia's Rubin-generation liquid-cooled data center design makes real strides on water consumption—but the full environmental math is a lot more complicated than the press release lets on.
Nvidia's new liquid cooling system genuinely eliminates on-site data center water use — but that only accounts for a quarter to a third of AI's total water footprint. The rest is hiding in the power grid.
Every ai hardware piece The Sideline has ever published.