Tesla is reviving its previously shelved Dojo3 project, aimed at creating a third-generation AI chip for electric vehicles. Now, it will be utilized for space computing, according to Elon Musk.
Now that the AI5 chip design is in good shape, Tesla will restart work on Dojo3.
— Elon Musk (@elonmusk) January 18, 2026
If you’re interested in working on what will be the highest volume chips in the world, send a note to AI_Chips@Tesla.com with 3 bullet points on the toughest technical problems you’ve solved.
Dojo is Tesla's specialized supercomputer and computing platform for training neural networks on vast amounts of video and sensor data.
The project was first mentioned in April 2019 when the company unveiled its autopilot chips. At that time, Musk positioned Dojo as a key component of Tesla's AI strategy and actively promoted the project in the following years.
The initiative faced challenges. In July 2024, ahead of the robotaxi presentation, Musk promised to "double down" on the platform's development. However, by August 2025, the billionaire unexpectedly announced the closure of the initiative and disbanding of the team. Despite plans to release a second-generation system (Dojo 2) in 2026, he labeled the project an "evolutionary dead end."
Now, more ambitious plans are being set for the development of Dojo.
AI4 by itself will achieve self-driving safety levels very far above human.
— Elon Musk (@elonmusk) January 18, 2026
AI5 will make the cars almost perfect and greatly enhance Optimus.
AI6 will be for Optimus and data centers.
AI7/Dojo3 will be space-based AI compute.
“AI7/Dojo3 will be designed for space-based AI computing,” Musk stated.
Tesla is preparing to reassemble the team that was disbanded a few months ago.
“If you’re interested in working on chips that will be the most widely produced in the world, send a note to AI_Chips@Tesla.com with three bullet points on the toughest technical problems you’ve solved,” the entrepreneur noted.
Space: The Future Trend
Musk and several other company leaders believe that the future of data centers lies beyond Earth. They argue that the planet's energy grids are nearing their limits.
Among the advantages are virtually unlimited access to solar energy and space for equipment placement. The downside is the high cost of launching rockets with the necessary infrastructure.
Analysts from the research group 33FG estimate that AI computing in orbit will become economically viable by 2030.
One of the first initiatives came from Google, which announced a plan to create a network of satellites in low Earth orbit to generate energy for powering data centers.
The idea is also supported by OpenAI CEO Sam Altman, but Musk has a strategic advantage—control over delivery systems.
The entrepreneur plans to use the upcoming SpaceX IPO to fund his idea of using Starship to launch groups of computing satellites that can operate in constant sunlight and collect energy around the clock.
Recall that in May 2025, China sent 12 satellites into space as part of a project to launch a network of orbital supercomputers.