Can We Trust AI to Keep Us Safe? Richard Hawkins and Philippa Ryan
Event details
AI is very clever at doing lots of things very well. Everyone is aware of self-driving cars carrying passengers around American cities; AI chat bots holding intelligent conversations with people; and machines that can identify the presence of cancer in scans with uncanny accuracy. But is it always doing so safely?
The real-world is complicated and unpredictable. So how can we trust the AI to behave safely all the time? Richard Hawkins and Philippa Ryan of the University of York look at examples of AI doing strange things and explore what’s going wrong and why. They’ll show you, from the important research they do at the Centre for Assuring Autonomy, how AI learns what to do from data and how small changes in that data can have big consequences.
Learn more about the Centre for Assuring Autonomy, a £12million partnership between Lloyd’s Register Foundation and the University of York.
About the speakers
Dr Richard Hawkins is a Senior Lecturer in the University of York’s Department of Computer Science. His research focuses on safety assurance of autonomous systems and AI. In his role at the Centre for Assuring Autonomy, he leads projects focused on developing safety cases for systems that incorporate machine learning and ensuring their ongoing safety in operation. Richard has collaborated with industry partners on numerous research projects and published extensively on the safety of software-intensive systems. Previously he has worked as a software safety engineer for BAE Systems and as a safety adviser for BNFL.
Dr Philippa Ryan is a Research Fellow in Assuring AI and Autonomous Systems at the University of York, and has over 20 years of experience from both academia and industry. Her research revolves around assuring through-life safety for complex and dynamic software including Autonomous Systems and Machine Learning. She has authored and co- authored dozens of papers about software safety in multiple applications, including Autonomy and AI. She previously worked as a principal consultant, acting as an independent software safety advisor and writing safety cases. She has worked in a range of sectors including civil avionics, maritime, medical devices, and automotive. She has successfully secured funding for, then led, multiple research and industrial projects, working with partners from diverse disciplines and backgrounds. She is chair of the Safety Critical Systems Club (SCSC) Safety of Autonomous Systems Working Group (SASWG).
Partners
.png)
