Weapons of Math Destruction

Weapons of Math Destruction examines how mathematical models and algorithms, often presented as objective and neutral, systematically disadvantage vulnerable populations. O'Neil, a data scientist and mathematician, argues that these "weapons of math destruction" (WMDs) operate at scale, lack transparency, and resist accountability—making them far more dangerous than individual human bias.

The book is structured around case studies across multiple domains: hiring algorithms that discriminate against women and minorities, predictive policing systems that concentrate enforcement in poor neighborhoods, credit scoring models that trap people in poverty, and educational algorithms that limit opportunity. O'Neil demonstrates how these systems are often poorly designed, trained on biased historical data, and deployed without adequate oversight or recourse for those harmed.

Published in 2016, the book became influential in sparking public and policy conversations about algorithmic fairness, transparency, and regulation. It helped establish a framework for thinking about algorithmic bias not as a technical glitch but as a systemic problem embedded in how organizations make decisions at scale. The work has been widely adopted in academic courses and cited in policy discussions around AI ethics and algorithmic accountability.