How AI became Seriously Useful and not a Threat to Humanity

Everyone thinks they know about artificial intelligence but not a lot of people know that many of the foundations of AI were built in Edinburgh. We’ve seen its birth, adolescence and now its difficult middle age at which it's become a big part of the community.  Unlike the early days, a lot of this stuff now works, although not always as straightforwardly as you might hope. In this talk, we'll cover the capabilities of AI and how it might mature, which is coloured by how it has grown up.

Poplualr media has gripped us with the idea of an AI system gone rogue. I-Robot. Ex-Machina. Age of Ultron. However, worrying about machines that are too smart, may be distracting us from the real threat: machines that are too dumb. AI systems must have the ability to explain their reasoning in a way that users can understand and assent to, we are responsible for them. Because of their open-ended behaviour, AI systems are also inherently hard to verify. Dr Alan Bundy will explain in this talk why we shouldn't worry about smart machines turning on us.