technology

AMD to acquire AI software startup as it seeks to catch up with Nvidia


Lisa Su, CEO, AMD

Scott Mlyn | CNBC

AMD announced Tuesday it will acquire Nod.AI, an open-source AI software startup, in order to expand the chipmaker’s AI software operations.

Nod.AI, or Nod Labs, builds open-source technologies “for future AI systems,” according to the startup, and it mainly specializes in reinforcement learning, a type of system that “learns” via trial and error.

The acquisition is part of AMD’s “AI growth strategy” that it hopes will better shore up its competition against rival chipmaker Nvidia, which saw its revenue spike more than 100% year over year for the quarter ended July 30, 2023. During the same time period, AMD reported a revenue decline of 18% but still beat analysts’ estimates for sales and profit.

“The acquisition of Nod.ai is expected to significantly enhance our ability to provide AI customers with open software that allows them to easily deploy highly performant AI models tuned for AMD hardware,” Vamsi Boppana, SVP of AMD’s AI group, said in a release.

Prior to founding Nod in 2013, Anush Elangovan, the startup’s CEO, was part of the first Chromebooks team at Google and a lead engineer at Cisco. Harsh Menon, Nod’s co-founder and CTO, previously worked at Kitty Hawk, the electric aircraft company backed by Google co-founder Larry Page, which shut down in 2022.

Nod Labs was initially billed as an AI hardware company focusing on gesture recognition and motion-tracking wearables, such as Bluetooth-connected rings for use in gaming. (Think: Making a real-life hand gesture using your thumb and forefinger to shoot a gun in a virtual reality game.) The startup raised millions in funding from VC firms like Menlo Ventures and Sequoia Capital, and team members reportedly hailed from Apple, Google, Samsung and more.

Readers Also Like:  Twitter says removed more content in first half of 2022 than in previous six months

AMD and Nod.AI did not respond to requests for comment.



READ SOURCE

This website uses cookies. By continuing to use this site, you accept our use of cookies.