Scientists shouldn't be afraid of risk

Humans are awful at calculations. In June I was at the Lindau Nobel laureate meeting, where Turing award winner Joseph Sifakis told me how to beat the Turing test, the way to distinguish between human and computer. ‘Just ask it to tell you pi to 9000 digits,’ he said. A human can’t do maths that quickly; a computer wouldn’t realise it was being tricked.

This is particularly true when it comes to risk. We have entire industries built up around our inability to compute odds – whether it’s gambling, insurance or risk-benefit profiles. While at Lindau, 2016 chemistry Nobel winner Jean-Pierre Sauvage told me his own thoughts on risk. An expert in inorganic photochemistry, Sauvage noticed that, increasingly, his work was drifting into a new, unexplored field: molecular machines. He took the leap and never looked back. ‘If you jump in it, if you are successful, you open up a new field.’