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.