A couple months ago we talked about some extravagant claims made by Google engineer Blaise Agüera y Arcas, who pointed toward the impressive behavior of a chatbot and argued that its activities “do amount to understanding, in any falsifiable sense.” … Continue reading …