Jobs at Alignment Research Center

Our current work focuses on developing an alignment strategy that could be adopted in industry today while scaling gracefully to future ML systems. Right now Paul Christiano, Mark Xu, and Beth Barnes are researchers and Kyle Scott handles operations. We believe that modern ML techniques would lead to severe misalignment if scaled up to large enough computers and datasets. Practitioners may be able to adapt before these failures have catastrophic consequences, but we could reduce the risk by adopting scalable methods further in advance. Our research priorities and methodology is described in our report on Eliciting Latent Knowledge (docs.google.com/document/d/1WwsnJQstPq91_Yh-Ch2XRL8H_EpsnjrC1dwZXR37PC8/edit?…). At a high level, we’re trying to figure out how to train ML systems to answer questions by straightforwardly “translating” their beliefs into natural language rather than by reasoning about what a human wants to hear. The evaluations project at the Alignment Research Center is a new team building capability evaluations (and in the future, alignment evaluations) for advanced ML models.