The Landscape of AI Safety and Beneficence Research - Richard Mallah. This FLI paper gives numerous examples of research topics aimed at ensuring that AI remains robust and beneficial, and was part of the Beneficial AI 2017 conference.