Even AlphaZero Found This Game Hard

Even Superhuman Go AIs Have Surprising Failure Modes — LessWrong

Alpha Zero Teaches Itself Chess 4 Hours, Then Beats Dad

Has the Alpha Zero chess program been made to play the Evans Gambit against itself, in an attempt to discover whether that gambit, with best play, is theoretically sound or whether White

Four hours after being taught the rules of chess, AlphaZero became the strongest player the world has ever seen • Graham Cluley

Why Artificial Intelligence Like AlphaZero Has Trouble With the Real World

Limits to Super-Intelligence. While the community struggles to sort…, by Kevin Zatloukal

DeepMind's AlphaZero Game Changer 1

Simple Alpha Zero

Assessing Game Balance with AlphaZero: Exploring Alternative Rule Sets in Chess (Paper Explained)

Google DeepMind's 'superhuman' AlphaZero computer taught ITSELF to beat humans at chess, shogi and Go

Alpha Zero and Monte Carlo Tree Search