We must regulate AI now to improve our lives and avoid its risks

DONE wisely, artificial intelligence "can be the best thing ever for humanity", says the fundamental physicist turned AI researcher Max Tegmark in our interview this week (see "If we do it wisely, AI can be the best thing ever for humanity"). We subscribe wholeheartedly to his assessment. Seldom has there been a technology with such an obvious power to improve our lot – or one with such obvious dangers. The risks are potentially existential.