AI safety courses and concepts

AI safety is the work of making sure AI systems do what they are meant to do and not something harmful — covering how models fail, how people misuse them, and the rules and controls placed around both. It is unusual among technical subjects in that its hardest problems are not technical: a system can work exactly as designed and still produce an outcome that is unfair, unexplainable, or illegal in the country it is running in. Learning it means holding two things at once — the specific ways models go wrong, and the governance built to catch them before anyone else does.