When it comes to chips for artificial intelligence, obviously the name that automatically comes to mind is Nvidia. The company is making a fortune selling semiconductors used for hot AI applications like large language models, and stock investors have rewarded it handsomely for doing so. But of course, Nvidia's GPUs can be used for more than just AI. They're also used for video games, graphics, cryptocurrency mining and more. But a new startup called MatX is aiming to build the ultimate chip just for LLMs. Co-founders Reiner Pope and Mike Gunter spent years at Alphabet, which has its own internal semiconductor operations, and now they've stalked out on their own to create a new chip company from scratch. We talk about how they're going about their job, what it takes to actually design and build a chip, and what it will take to get customers to switch over from the industry leader.

For more episodes on this topic:
Coreweave's CSO on What It Really Takes to Build an AI Datacenter

Inside the Battle for Chips That Will Power Artificial Intelligence

Only Bloomberg.com subscribers can get the Odd Lots newsletter in their inbox each week, plus unlimited access to the site and app. Subscribe here: 

See omnystudio.com/listener for privacy information.