A large and varied cross-section of academics, students, practitioners and policy makers made up the audience at the The Centre for Information Rights 4th Winchester Conference on Trust, Risk, Information and the Law on Wednesday 3rd May at the University of Winchester.
The day started with a keynote speech from Professor Katie Atkinson from the University of Liverpool focusing on how Artificial Intelligence can be used in supporting legal practice, with an engaging example of whether an AI model she'd help create could predict the rules of a famous US case involving the disputed ownership of a baseball. At present, the model is currently being tested to see if it can predict the outcomes of decided legal cases but has the potential for much greater use within the legal field. On a related topic, speakers from Durham Constabulary discussed their research into using AI to make decisions about granting bail to offenders – a talk that caught the eye of the Law Society Gazette.
Further sessions were held throughout the day on topics ranging from autonomous weapons to "fake news" to smart contracts to whether current legal principles could be used to apply to new and untested legal issues arising from the use of AI in daily life; with a particularly healthy debate on liability for autonomous vehicles.
The afternoon keynote speaker, John McNamara a Senior Inventor at IBM, got the audience guessing at what his "Probey Mk 2" had been getting up to (the answer being fired to the edge of space and seeing if it had learned what data and images it was meant to be capturing were actually caught!) while impressing on the audience the need to protect trust in a world disrupted by human learning. He argued that machine learning systems should behave according to human values and be aligned o the environment they will function in, with the core theme of "trust" flowing into the final panel discussion of the day chaired by Timandra Harkness. A wide ranging discussion between the panel members and audience on ethics, risk, politics and the ability of law and regulators to deal with the ever expanding use and scope of AI brought a stimulating and thought-provoking end to the day.