Evaluating the ethics of autonomous systems MIT researchers developed a testing framework that pinpoints situations where AI decision-support systems are not treating people and communities fairly. Published: 2026-04-02