Vanarama analyzed the processing power of Tesla’s microchips and found that Tesla’s microchips will surpass the human brain (10 trillion operations per second) in 11 years (i.e., 2033). In addition, Tesla’s D1 chip is 30 times more powerful than the chip they used 6 years ago.

At last year’s Tesla AI Day, Tesla unveiled its self-developed neural network training chip, the D1 chip, which is manufactured using TSMC’s 7nm process and has a core area of 645 square millimeters, integrating up to 50 billion transistors, 354 training nodes, and 17.7 kilometers of internal circuitry alone. The core area is 645 square millimeters, integrating up to 50 billion transistors, 354 training nodes, and 17.7 kilometers of internal circuitry alone, achieving superb computing power and ultra-high bandwidth.

In early June, Tesla CEO Elon Musk posted that the second Tesla AI Day event would be postponed from August 19 to September 30, when the company may unveil a prototype of the humanoid robot Optimus.

In mid-August this year, Musk claimed that Tesla’s humanoid robot could be cheaper than a car in the future. In addition, he said that the robot is close to the height and weight of an adult, can carry or carry heavy objects, can also walk in small steps, and the screen on its face is an interactive interface for communication with people.

Musk has tweeted that the purpose of Tesla’s AI Day event is to convince talented artificial intelligence, software or chip talent to join Tesla.

The post Tesla microchip processing power will surpass the human brain by 2033 appeared first on TechGoing.