(Tesla’s phone)

This move comes just five months after Tesla suspended the Dojo project. Previously, after the departure of project leader Peter Bannon, Tesla disbanded the team responsible for the Dojo supercomputer. About 20 former team members subsequently joined DensityAI, an emerging AI infrastructure company co founded by former Dojo leader Gannis Venkataraman and former Tesla employees Bill Zhang and Ben Florin.

When the Dojo project was suspended, there were reports that Tesla planned to reduce its investment in self-developed chips and instead increase its reliance on computing resources from partners such as Nvidia and AMD, and chose Samsung to be responsible for chip manufacturing. Musk’s latest statement indicates that the company’s strategy may be adjusted again.

The AI5 chip currently used by Tesla is produced by TSMC and is mainly used to support autonomous driving functions and Optimus humanoid robots. Last summer, Tesla signed a $16.5 billion agreement with Samsung to produce the next generation AI6 chip, which will serve high-performance AI training in Tesla vehicles, Optimus robots, and data centers.

AI7/Dojo3 will focus on space AI computing, “Musk said on Sunday, meaning that the restarted project will be given a more cutting-edge positioning. To achieve this goal, Tesla is working on rebuilding the team that disbanded several months ago. Musk directly issued a talent recruitment invitation on the same occasion: “If you are interested in participating in the construction of the world’s most widely used chip, please feel free to send an email to AI_Chips@Tesla.com That’s right.

Roger Luo stated:Tesla’s restart of the Dojo3 towards space computing demonstrates its continuous exploration and rapid adjustment capabilities in AI chip strategy. This is not only a significant shift in its technological roadmap, but also reflects its early layout for future high frontier AI computing scenarios.

All articles and pictures are from the Internet. If there are any copyright issues, please contact us in time to delete.

Inquiry us [contact-form-7]

(Samsung Electronics Develops Neuromorphic Chip Prototype)

The chip uses a network of artificial neurons and synapses. It processes data in parallel, similar to biological systems. Traditional AI chips rely on sequential computations, which demand high power and generate heat. Samsung’s approach cuts energy consumption significantly. Tests show the prototype operates at a fraction of the power required by conventional hardware.

Engineers combined advanced memory technologies with novel circuit designs. This allows real-time learning and adaptation without external software updates. The chip could enable devices to respond dynamically to user behavior. Potential applications include robotics, smart sensors, and medical diagnostics.

Samsung emphasized the chip’s scalability. It can be integrated into small devices or expanded for data centers. The company plans to collaborate with global researchers to refine the technology. Partnerships will focus on optimizing algorithms and expanding use cases.

Industry experts note neuromorphic computing could address current AI limitations. Tasks like pattern recognition and decision-making may become faster and more accurate. Samsung’s prototype processes visual data 50 times quicker than existing chips in lab trials.

The project began in 2019 at Samsung’s Advanced Institute of Technology. Progress accelerated after breakthroughs in materials science and neural network modeling. Challenges remain in mass production and software compatibility. Engineers are working to standardize the design for commercial adoption.

(Samsung Electronics Develops Neuromorphic Chip Prototype)

A Samsung vice president stated the chip represents a shift toward bio-inspired computing. He highlighted sustainability benefits, citing reduced data center energy costs. The prototype is undergoing testing with select partners. No timeline for market release was provided.