Ai Safety is the field dedicated to ensuring that advanced Artificial Intelligence systems benefit humanity, rather than causing harm. It explores methods to build AI that is robust, trustworthy, and aligned with human values, a challenge often termed the alignment problem.