Résumé du livre
Cathy O’Neil has a PhD in mathematics from Harvard and was a teacher at Barnard College before moving to the private sector as a data scientist for various start-ups. Her writing can be found on the popular blog Mathbabe. Her other books include Doing Data Science.
The internet has proven to be a double-edged sword for democracy, amplifying diverse voices while being vulnerable to manipulation through algorithms that can subtly shape public opinion, as seen in voter influence during elections in the US and India. Predictive algorithms, similarly, have found their way into law enforcement, where they often perpetuate biases by disproportionately targeting impoverished neighborhoods and misclassifying individuals like Robert McDaniel as threats based solely on their environment and associations. In higher education, the rise of university rankings by *US News and World Report* inadvertently drove tuition costs up by 500% between 1985 and 2013, as institutions prioritized metrics over accessibility, harming students and schools alike. Employment practices have also been affected by data-driven tools, with cases like Kyle Behm’s rejection from jobs due to personality tests and Catherine Taylor’s wrongful association with a crime due to data errors, highlighting the flaws in such systems. Similarly, insurance companies use algorithms that often penalize financially disadvantaged individuals, such as drivers with poor credit but clean records, while favoring wealthier clients, further deepening inequalities. These examples underscore the unintended consequences of relying on algorithms across various sectors, setting the stage for the next discussion on their broader societal impact.
Pour lire le reste du livre, téléchargez
Bitely