One area where AI has been used, with some disastrous results, is in police use of facial recognition neural net programs. The leading case comes from Detroit in the United States, where a Black man, Robert Williams was arrested for robberies, based upon surveillance footage. It turned out that Williams was innocent, having an alibi. But he was still arrested on the basis of unclear grainy footage. It is known that there are limitations of the programs based upon the training data used for the neural net, which has been largely based upon white faces and creates a racial bias, with Black’s being 100 times more likely to be falsely identified than Whites. At present there is a movement to either ban or restrict the use of these technologies which conduct one-many matches of suspects faces to data banks such as licence face photos.