CHAI researcher Rohin Shah wrote a detailed review of public work in AI alignment in 2019 on the AI Alignment Forum. The review features work on topics such as AI risk analysis, value learning, robustness, and field building.