A talk and discussion about deepfakes, how they are generated, their various types, the risks they pose and that trying to make models safer by censoring their training data may not always be the best idea.
My research interests include distributed robotics, mobile computing and programmable matter.