- Tesla is spending $500 million to install a Dojo supercomputer at its Riverbend Gigafactory in Buffalo, New York, Gov. Kathy Hochul said in a press release Friday.
- Tesla’s Dojo supercomputer, which will be installed over the next five years, will be used to train its artificial intelligence systems that support autonomous driving.
- The Dojo supercomputer is expected to be one of the most powerful computing platforms in the world, with the ability to train machine learning models and process millions of terabytes of data from Tesla’s electric vehicles.
The data processed by the Dojo supercomputer will help Tesla improve the safety and engineering of its Full Self Driving and Autopilot autonomous driving features and advanced driver assistance systems, according to the press release.
During Tesla’s Q2 2023 earnings call last July, the company identified four main technology pillars required to enable vehicle autonomy at scale: large, real-world datasets; neural net training; vehicle hardware; and vehicle software. The Dojo supercomputer will support faster and cheaper neural net training for Tesla’s self-driving vehicles, according to the automaker.
“The better the neural net training capacity, the greater the opportunity for our Autopilot team to iterate on new solutions,” Tesla wrote in its Q2 2023 earnings presentation.
In September, Morgan Stanley analysts led by Adam Jonas wrote in a note that Tesla’s Dojo supercomputer could boost the electric automaker’s market value by $600 billion by supporting AI-powered robotaxis, vehicle software, and advanced computer vision processing, according to a Reuters report.
Dojo can open up new addressable markets that "extend well beyond selling vehicles at a fixed price," the analysts wrote. "If Dojo can help make cars 'see' and 'react,' what other markets could open up? Think of any device at the edge with a camera that makes real-time decisions based on its visual field."
In December, Bloomberg reported that Tesla’s Dojo project leader Ganesh Venkataramanan departed the company. Former Apple Inc. executive and Tesla director Peter Bannon leads the project now.