Assessing AI's progress toward replicating the full breadth and depth of human intelligence is a deceptively hard problem. A paper by François Chollet, who was

Assessing AI's progress toward replicating the full breadth and depth of human intelligence is a deceptively hard problem. A paper by François Chollet, who was until recently a researcher at Google, called "On the Measure of Intelligence" is the best overview of the challenges I've read. Highly recommended. https://arxiv.org/abs/1911.01547 https://arxiv.org/abs/1911.01547