In Human Compatible, his new book on artificial intelligence (AI), Stuart Russell confronts full on what he calls "the problem of control". That is, the possibility that general-purpose AI will ultimately eclipse the intellectual capacities of its creators, to irreversible dystopian effect. The control problem is not new. But, by 1950, Norbert Wiener, the inventor of cybernetics, was writing (in The Human Use of Human Beings) that the danger to society "is not from the machine itself but from what man makes of it". Russell's book in effect hangs on this tension: whether the problem is controlling the creature, or the creator.

Only one of them is what we might think of as a tech guy, and it's that difference that means the other is likely to be right. Musk, a physicist by training, is a well-known AI radical who sees the technology as a threat to the human race because, in his view, it will inevitably outsmart us and start running the world without heeding our needs. "The biggest mistake I see AI researchers making is assuming that they're intelligent," he said during a debate in Shanghai on Thursday. "They're not, compared to AI." He likened humanity to a bootloader – a small piece of software needed to turn on a computer.

Tech billionaires Jack Ma and Elon Musk can't agree whether artificial intelligence is going to take over the world.