Weapons of Math Destruction: How Big Data Increases Inequality and Threatens Democracy

(Read time: 1.5 - 2 Minutes)

By Ollie Coull

Weapons of Math Destruction by Cathy O'Neil is the perfect book to review if you want to talk about how the law is completely failing to keep up with technology. Personally, I would give it a solid 8/10. It’s an absolute eye-opener of a book, but it does have a couple of flaws that keep it from being a perfect 10.

Going into it, I thought algorithms and computer code were just neutral formulas used to make life more efficient. This book completely shattered that idea and earned most of its points by showing how math is low-key being used to ruin people's lives. O'Neil explains how big data models, which she calls "Weapons of Math Destruction" (WMDs), are totally secret, affect millions of people, and are impossible to fight back against.

The justification for the high rating comes down to how well she breaks down these "toxic loops." She shows how algorithms decide who gets a car loan, who gets hired, or even how police deploy officers. Because these programs are trained on historical data, they just copy all the biases and racism of the past, wrap it up in a "scientific" computer program, and call it objective. If an algorithm flags you as a risk because of your zip code, you get denied a loan, which keeps you poor, which feeds back into the algorithm to prove it was "right." For a law blog, it gives you a ton of concrete examples to argue why tech needs way more legal regulation, especially when companies hide behind "trade secrets" to keep their code from being scrutinized in court.

However, the reason it drops two points down to an 8 is that the book gets pretty repetitive after the halfway mark. Once you understand her main point in the first few chapters, the rest of the book just applies that exact same formula to different industries over and over again—education, workplace scheduling, insurance—without really adding anything new. Plus, it’s a total downer. She is amazing at pointing out everything that is legally and ethically broken, but she doesn't offer a ton of realistic solutions on how lawmakers or programmers can actually fix it.