Some outrageously unnecessary, and popular story of software, particularly #AI catches your attention. You trace the references and read the original paper "Multi-agent Reinforcement Learning in Sequential Social Dilemmas" (https://arxiv.org/abs/1702.0303