Tesla's CEO Elon Musk has long touted Dojo as the cornerstone of the company's artificial intelligence (AI) ambitions, but what exactly is Dojo, and why is it so critical to Tesla's goal of achieving full autonomy in its vehicles? In short, Dojo is Tesla's custom-built supercomputer designed to train its "Full Self-Driving" (FSD) neural networks, which are currently used in hundreds of thousands of Tesla vehicles. The beefing up of Dojo goes hand-in-hand with Tesla's goal to reach full self-driving and bring a robotaxi to market.

Tesla's FSD system, which currently costs $8,000 and has been priced as high as $15,000, can perform some automated driving tasks but still requires a human to be attentive behind the wheel. The company plans to launch an unsupervised FSD for U.S. customers in 2025. Meanwhile, Tesla is gearing up to launch an autonomous ride-hail service using its own fleet of vehicles in Austin this June. To achieve this, Tesla needs to store and process vast amounts of video data collected from its cars around the world and run millions of simulations to train its model on the data.

That's where Dojo comes in – a supercomputer system designed to function as a training ground for AI, specifically FSD. The name is a nod to the space where martial arts are practiced. A supercomputer is made up of thousands of smaller computers called nodes, each with its own CPU and GPU. The former handles overall management of the node, and the latter does the complex tasks, like splitting tasks into multiple parts and working on them simultaneously. GPUs are essential for machine learning operations like those that power FSD training in simulation.

Tesla's vision-only approach is the main reason the company needs a supercomputer. The neural networks behind FSD are trained on vast amounts of driving data to recognize and classify objects around the vehicle and then make driving decisions. That means that when FSD is engaged, the neural nets have to collect and process visual data continuously at speeds that match the depth and velocity recognition capabilities of a human.

Tesla is working on a next-gen D2 chip that aims to solve information flow bottlenecks. Instead of connecting individual chips, the D2 would put the entire Dojo tile onto a single wafer of silicon. The company also hopes to make something better that increases bandwidth and decreases latencies. That's why the automaker's AI division decided to come up with its own custom hardware program that aims to train AI models more efficiently than traditional systems.

Taking control of its own chip production means that Tesla might one day be able to quickly add large amounts of compute power to AI training programs at a low cost, particularly as Tesla and TSMC scale up chip production. It also means that Tesla may not have to rely on Nvidia's chips in the future, which are increasingly expensive and hard to secure.

In the long run, Tesla could theoretically create a new business model based on its AI division. Musk has said that the first version of Dojo will be tailored for Tesla computer vision labeling and training, which is great for FSD and for training Optimus, Tesla's humanoid robot. But it wouldn't be useful for much else.

Musk has said that future versions of Dojo will be more tailored to general-purpose AI training. One potential problem with that is almost all AI software out there has been written to work with GPUs. Using Dojo to train general-purpose AI models would require rewriting the software.

A September 2023 report from Morgan Stanley predicted that Dojo could add $500 billion to Tesla's market value by unlocking new revenue streams in the form of robotaxis and software services.

In short, Dojo's chips are an insurance policy for the automaker, but one that could pay dividends. As Tesla continues to scale up its AI ambitions, the success of Dojo will be critical to achieving its autonomy goals.