The Open Language Safety Research (OpenLSR) project is a series of research efforts led by MIT CSAIL aiming to improve the robustness, inclusion, and alignment of language and conversation AI.