Breaking News
The Absolute Farce of Earth Day 2026
Trump On Iran: 'Lots Of Bombs Will Go Off' If No Agreement
I Tested the Top 7 Salts for Toxins (Only 2 Passed)
Is it possible to NOT pay federal income taxes legally?
Top Tech News
Researchers Turn Car Battery Acid and Plastic Waste into Clean Hydrogen and New Plastic
'Spin-flip' system pushes solar cell energy conversion efficiency past 100%
A Startup Has Been Quietly Pitching Cloned Human Bodies to Transfer Your Brain Into
DEYE 215kWh LiFePO4 + 125,000W Inverter + 200,000W MPPT = Run A Factory Offgrid!!
China's Unitree Unveils Robot With "Human-Like Physique" That Can Outrun Most People
This $200 Black Shaft Air Conditions Your Home For Free Forever -- Why Is It Banned in the U.S.?
Engineers have developed a material capable of self-repairing more than 1,000 times,...
They bypassed the eye entirely.
The Most Dangerous Race on Earth Isn't Nuclear - It's Quantum.
This Plasma Stove Cooks Hotter Than The Sun 
Tesla Dojo, Supercomputers and AI
There does not seem to be a limit for neural nets to utilize more resources to get better and faster results.
Tesla is motivated to develop bigger, faster computers that are precisely suited to their needs.
The Google TPU architecture has not evolved as much over the last 5 years. The Google TPU chip is designed for the problems that Google runs. They are not optimized for training AI.
Tesla has rethought the problem of AI training and designed the Dojo AI supercomputer to optimally solve their problems.
If Tesla commercializes the AI supercomputer that will help to get to lower costs and greater power with more economies of scale.
One of the reasons that TSMC overtook Intel was that TSMC was making most of the ARM chips for cellphones. TSMC having more volume let them learn faster and drive down costs and accelerate technology.
99% of what neural network nodes do are 8 by 8 matrix multiply and 1% that is more like a general computer. Tesla created a superscalar GPU to optimize for this compute load.