Tesla building an AI chip for its cars with AMD, GlobalFoundries

Tesla is getting closer to having its own chip for handling autonomous driving tasks in its cars.

The carmaker has received back samples of the first implementation of its processor and is now running tests on it, said a source familiar with the matter.

The effort to build its own chip is in line with Tesla’s push to be vertically integrated and decrease reliance on other companies.

But Tesla isn’t completely going it alone in chip development, according to the source, and will build on top of AMD intellectual property.

AMD shares spiked after CNBC reported that the company is working with Tesla. Shares of the stock ended the day nearly 5 percent higher and continued to climb after hours.

On Wednesday Sanjay Jha, CEO of AMD spin-off GlobalFoundries, said at the company’s technology conference in Santa Clara, California, that the company is working directly with Tesla. GlobalFoundries, which fabricates chips, has a wafer supply agreement in place with AMD through 2020.

A more power-efficient purpose-built chip could help Tesla get closer to delivering totally autonomous driving. Tesla CEO Elon Musk promised this year that capability will be available to consumers in 2019.

Tweet

Tesla’s silicon project is bounding ahead under the leadership of longtime chip architect Jim Keller, the head of Autopilot hardware and software since the departure of Apple veteran Chris Lattner in June. Keller, 57, joined Tesla in early 2016 following two stints at AMD and one at Apple. Keller arrived at Apple in 2008 through its acquisition of Palo Alto Semiconductor and was the designer of Apple’s A4 and A5 iPhone chips, among other things.

More than 50 people are working on the initiative under Keller, the source said. Tesla has brought on several AMD veterans after hiring Keller, including director Ganesh Venkataramanan, principal hardware engineer Bill McGee and system circuit design lead Dan Bailey.

AMD and Tesla did not immediately respond to requests for comment.

Current Tesla vehicles use Nvidia graphics processing units as part of the Autopilot self-driving hardware. Previously Tesla used chips from Mobileye, but the two companies parted ways last year after an accident involving a Mobileye-powered Tesla. Earlier this year Intel acquired Mobileye for more than $15 billion.

Notably Tesla is not working with Intel, which has enjoyed a long working relationship with Alphabet’s Waymo car group.

As a rule, GPUs from the likes of Nvidia, which are widely used today for various types of AI workloads, are capable of doing many things but aren’t designed with just a few narrow computing jobs in mind, hence the custom chip development. Plus, Tesla would be less impacted by pricing dictated by Nvidia if it switches to in-house hardware.

Alphabet has already built two generations of an AI chip that can work as a GPU alternative. Microsoft is working on its own AI chip for the HoloLens. And Apple has incorporated an AI chip into its top-of-the-line iPhone X.