Its been revealed in The Telegraph that driverless cars produced in the US are unable to identify vehicles such as buses and taxis, and therefore pose a significant risk on Britains roads.
Engineers have identified a serious flaw within the artificial intelligence, in so far as the vehicle's software has only been "taught" to identify vehicles that are used in the U.S.
This has now caused serious concerns among scientists and politicians in the U.K. who are now questioning companies who are developing autonomous technology, in relation as to how safe the vehicles can be on the U.K's road network.
Companies such as Google, Uber, as well as others, who are forging ahead with driverless vehicle technology have been accused of developing artificial intelligence software which is only capable of being used safely within the U.S. market, whilst the rest of the world's vehicular customs and culture are being sidestepped.
There are a number of MPs in the U.K. calling for an enquiry into the issue before any U.S. manufactured autonomous vehicle is allowed on British roads. It is believed that the only way to overcome this issue is to identify every single vehicle on British roads as well as identifying them in every kind of weather condition. This therefore begs the question, if the vehicles cannot recognise basic vehicles such as buses and taxis, how will they identify vehicles which have modifications that aren't attributed to any given manufacturer, how will kit cars be recognised, how will various types of mobility scooter, wheelchair, even pram be recognised? There seems to be no explanation as to how the vehicles will be updated when new models of any of the above emerge, nor any explanation as to what happens if an update fails. Expanding that further, if an update on an autonomous vehicle fails and an accident occurs, who is responsible? There seems to be more questions than answers at the moment, as we await to hear from Silicon Valley as to how these issues will be addressed.