A global hacker 'red team' and rewards for hunting algorithmic biases are just some of the recommendations from experts who argue that AI faces a 'tech-lash' unless firm measures are taken to increase public trust.