Why chips?

Chip technology and hardware advances are the foundational basis for advancements in AI. 

Machine learning is a computational process... computational power and computing architectures shape the speed of training and inference in machine learning, and therefore influence the rate of progress in the technology.
— Tim Hwang, MIT Media Lab/Google

There is a diverse array of chip technology and types that have contributed to the flowering in AI. In particular, GPUs made by NVIDIA have been making huge improvements over the long-existing CPU-driven computing stack.

 

GPUs have flowered since they enable more parallel computation than CPUs. This parallel computation is key to processing the vast number of inputs AI systems need to receive.

CPUs vs. GPUs
 

While GPUs have been remarkably successful in a short time frame, they were not originally built with the advances of AI in mind--they were made for gaming!

While GPUs are good at linear algebra, their lead is being challenged by dozens of Chinese and American companies creating chips designed from the ground up for linear algebra computations.
— Reza Zadeh, Stanford University