A senior research scientist at DeepMind, who specialized in robotics and artificial intelligence, has departed from Google to establish a new robotics startup named Generalist AI. This venture has already secured investment from Nvidia, according to information obtained by TechCrunch.
Pete Florence was identified as both co-founder and CEO of Generalist AI during a panel held at Nvidia’s GTC conference in San Jose recently. The panel included portfolio companies associated with Nvidia’s venture capital branch, NVentures.
NVentures has taken an increasingly active role as a venture capital firm following Nvidia’s significant financial success in the burgeoning AI sector.
“We are largely still in stealth,” Florence shared with TechCrunch, elaborating that the startup’s mission is “to make general-purpose robots a reality.” Florence’s departure from DeepMind took place a year ago, as indicated on his LinkedIn profile. Kamyar Ghasemipour, a student researcher at DeepMind, has also joined Generalist AI as a founding technical staff member, according to his LinkedIn details.
Florence is among several former DeepMind employees who have launched their own companies, including Reflection AI, an autonomous coding startup, Latent Labs, a biotech startup, and Mistral, among others.
In addition, within Alphabet but outside of DeepMind, prominent leaders responsible for Google’s successful NotebookLM product left to start their own AI company late last year.
DeepMind’s robotics division recently introduced a series of new AI models designed for robot control, a development documented in a paper that cites four works co-authored by Florence.
Although Florence refrained from detailing specific aspects of his startup during the GTC panel, it is evident that the focus will center on robotics.
“We are dead set on making robots that can do absolutely anything,” Florence stated in response to a query about the potential impacts if his startup achieved substantial success.
He added, “So just imagine a world where the marginal cost of physical labor is driven to zero.”
Nvidia chose not to comment, and Google has not responded to requests for comment.
