Google CEO Sundar Pichai’s solution for a shortage of people with machine learning skills is relatively simple: just let its machine learning system do the work of creating newer, better machine learning systems.
The Google Brain research division, which is dedicated to machine learning, is already seeing better results from the machines than its human experts in some areas.
But if we have already reached a point where much of the processes that AI carry out are difficult to understand, what will happen when these same machines start building their own AI?
This comes as Google announces an AI dedicated processing chip and new supercomputer, harnessing the power of TensorFlow.
Image: Pastiche artwork created by Google Brain
This story is taken from the 19 May 2017 edition of The Warren Centre’s Prototype newsletter. Sign up for the Prototype here.