> I'm arguing that animal or lesser intelligence is built around hundreds of thousands of parlor tricks operating in a complex ensemble. Until ML/AI can perfor

> I'm arguing that animal or lesser intelligence is built around hundreds of thousands of parlor tricks operating in a complex ensemble. Until ML/AI can perform a single one of those parlor tricks without the constant direction of human intelligence, there’s no reason to stop marveling.