Tesla news: Company employs former AMD staff help for their own AI chips

Elon Musk promoting his Tesla car modelsREUTERS/Stephen Lam

It seems Elon Musk is once again setting his sights on another milestone for humankind, as the CEO of Tesla has now announced that his company is now working on its own artificial intelligence (AI).

Tesla is now getting help from former Advanced Micro Devices (AMD) employee Jim Keller to develop its own AI chips to power their own proprietary AI technology. Keller is best known for his work on K8, K12, and ZEN central processing unit (CPU) microarchitectures, as well as the instruction set for x86-64 operating systems.

Keller has since left AMD and joined Tesla as the Vice President of Autopilot Hardware Engineering, meaning he is now working on the Tesla cars instead of CPUs. Musk has confirmed at the Neural Information Processing Systems (NIPS) conference that Keller was spearheading Tesla's development for the said AI chip.

"I want to make it clear that Tesla is serious about artificial intelligence, such as software and hardware, and we are developing our own AI chips," said Musk. Website Guru3D has stated that Musk now pursuing AI is quite a curious case since he has previously warned people of the danger of the said technology.

Because of this new venture by Musk and Tesla, this will pit him against the current leading AI developer company, which is graphics processing unit (GPU) manufacturer NVIDIA. Tesla is confident that it can keep up with the AI race, especially with Keller having a rumored team of fifty people focusing on the development of the technology.

Tesla's first application for their AI is expected to be for their own brand of automobiles, where the AI must process a large amount of swiftly-generated data such as road objects or directions. These must both be analyzed swiftly and when successfully developed and implemented, might eliminate the need for a human driver.

Currently, though, Tesla will have to make do with NVIDIA chips for the AI of their car, that is until they make their own AI chips on par or better with those made by the GPU company.