Fazl Barez of the University of Oxford queries how artificial intelligence built to serve a better purpose has the potential to be dangerous in the wrong hands.
Read more: If AI robots can be tricked into ‘going rogue’, what are the implications?