Clarendon Laboratory, Department of Physics, University of Oxford, Parks Road, Oxford, OX1 3PU
Dr Anders Sandberg, Institute for Futures Studies, Stockholm
Joy Blanchard at tpadmin@physics.ox.ac.uk
What if we succeed? Artificial intelligence past, present and future
Abstract: This talk is a brief reflection on where artificial intelligence comes from, where it appears to stand, and were it may be going, with an eye on its link to physics. AI has a long history associating closely with physics, from cybernetics over pattern recognition for particle physics to the deep connections in statistical learning theory that made neural networks what they are today. At present it both acts as an enabler for new research capabilities and posing challenging questions in the borderlands of computation and physics. It also poses important challenges in safety and what the future of scientific discovery (and academic education/research) will be.