Also called: Alignment Researcher, Interpretability Researcher
Studies how to make powerful AI systems honest, controllable, and safe.
You might peer inside a neural network to find which parts represent a concept, or design tests to see whether a model would try to deceive its users. It's part computer science, part philosophy, and taken very seriously at AI labs.
Estimate; frontier labs pay very well; nonprofits pay less.
Approximate: anchored on Bureau of Labor Statistics data for a broader job group.
Doctorate (PhD), about 8 years after high school. Often a PhD, though some labs hire strong engineers with bachelor's degrees.
Ways in: 4-year college, Graduate school, Self-taught. Training cost: $15k – $60k.
Majors to look at: Computer Science, Mathematics, Philosophy, Cognitive Science.
Growing much faster than average. Small but fast-growing field.
Low risk in 5 years, medium in 10. Need for safety work grows with AI power; AI may assist with experiments, but setting goals and judging risks stays human.
Classes: AP Calculus, AP Computer Science A, Philosophy or ethics, AP Statistics.
Official data: O*NET occupation 15-1221.
Career Universe is an early version. Pay, training and AI-risk figures are estimates to start your own research, not verified facts. The research was written with AI research assistants, checked by validation scripts and spot-checked by hand. Check anything important with schools, employers and official sources. How the numbers work.