A talk that I gave about using argumentation-based frameworks to implement reward functions and implement (moral) values in AI systems, at the Value Connection workshop. It is targeted for a multidisciplinary audience.