Tesla, AMD Developing Self-Driving Car AI Chip

Tesla, AMD Developing Self-Driving Car AI Chip

It's no secret that Tesla always wanted to develop an in-house chip to handle its cars' autonomous functions.

Over 50 workers are developing the project under the leadership of Keller, including Bill McGee who is Principal Hardware Engineer at AMD and Dan Bailey, the system circuit design leader according to CNBC.

Sanjay Jha, CEO of GlobalFoundries, a spin-off of AMD, spilled the beans at its conference in Santa Clara, California, CNBC reports. Tesla has apparently already received samples of the chip and is running tests.

After hiring Andrej Karpathy as its new head of AI and Autopilot Vision, earlier this year, auto maker Tesla hopes to make a mark in the autonomous vehicle market and compete with the likes of Google and Samsung. Some analysts predict great things for the autonomous vehicle market, which is why Nvidia has been investing heavily in the field with devices like its Xavier system-on-chip (SoC) and the 2013 opening of a dedicated automotive technology centre and Intel has been hot on its heels with acquisitions including computer vision specialist Mobileye earlier this year. The desire to build its own chip is said to be aligned with Tesla's wish to be more vertically integrated and less reliant on other companies.

"It's a move that would clearly be in line with the recent hiring of ex-AMD executive Jim Keller to replace Chris Lattner as head of Tesla's Autopilot hardware engineering effort", reports The Verge. Neither AMD nor Tesla has commented upon this story either, so far. Keller previously worked at AMD and Apple, particularly on the latter's A4 and A5 chips.

There's no official comment yet from either AMD or Tesla, so we'll just have to wait and see whether this development means we're still on track for snooze-driving by 2019.