

š§ Listened in audio š¢ Narrated by Cathy OāNeil ā± Duration: 6 hours š·ļø Publisher: Random House Audio & Crown Publishing
Every AI and tech book I'd picked up in the last year had Cathy O'Neil's name in the footnotes, whispering go read this, it explains everything. This is the book everyone name-drops when talking about AI, algorithms, and the dark side of data. O'Neil delivers exactly what the references promised. The argument is airtight. The examples are damning. The picture she paints of how big data and algorithmic bias quietly wreck real lives is thorough, well-researched, and frankly a little terrifying. If you've ever wondered how a zip code can tank your credit score or how a personality test filters out job applicants before a human even glances at the pile, this book explains the machinery behind it all.
Here's my honest confession though: I DNF'd at 40%. Iām not mathematically inclined, and while OāNeil does work hard at analogy and explanation, the book still leans more āmath and modeling primerā than narrative nonfiction in places. I found myself disconnecting as she dug into the mechanics of models and feedback loops, even though I was fully on board with the message about inequality and systemic harm. The technical framing that makes it credible and rigorous is the exact same thing that kept pulling me out of it. I'd find myself re-listening to paragraphs, losing the thread, and zoning out somewhere between the regression models and the recidivism scores.
This is a sit-down-with-a-highlighter kind of book, and I gave it a walking-around-doing-laundry kind of attention. That mismatch is on me. Still, I think readers who are analytically inclined, or who are deep in the AI/tech policy space, will find this essential, not just interesting. I can absolutely see why this is a foundational text in the āAI harms / big data ethicsā conversation, but for my personal taste and attention span, it turned into more homework than compelling nonfiction.
Would I recommend it? If you're already fluent in data science, tech ethics, or policy, absolutely, non-negotiable, add it to your shelf immediately. Iād recommend Weapons of Math Destruction to readers who are comfortable with more technical explanations and want a foundational, early look at how algorithms scale inequality. For me, this landed as a respectful DNF at 40%.
š§ Listened in audio š¢ Narrated by Cathy OāNeil ā± Duration: 6 hours š·ļø Publisher: Random House Audio & Crown Publishing
Every AI and tech book I'd picked up in the last year had Cathy O'Neil's name in the footnotes, whispering go read this, it explains everything. This is the book everyone name-drops when talking about AI, algorithms, and the dark side of data. O'Neil delivers exactly what the references promised. The argument is airtight. The examples are damning. The picture she paints of how big data and algorithmic bias quietly wreck real lives is thorough, well-researched, and frankly a little terrifying. If you've ever wondered how a zip code can tank your credit score or how a personality test filters out job applicants before a human even glances at the pile, this book explains the machinery behind it all.
Here's my honest confession though: I DNF'd at 40%. Iām not mathematically inclined, and while OāNeil does work hard at analogy and explanation, the book still leans more āmath and modeling primerā than narrative nonfiction in places. I found myself disconnecting as she dug into the mechanics of models and feedback loops, even though I was fully on board with the message about inequality and systemic harm. The technical framing that makes it credible and rigorous is the exact same thing that kept pulling me out of it. I'd find myself re-listening to paragraphs, losing the thread, and zoning out somewhere between the regression models and the recidivism scores.
This is a sit-down-with-a-highlighter kind of book, and I gave it a walking-around-doing-laundry kind of attention. That mismatch is on me. Still, I think readers who are analytically inclined, or who are deep in the AI/tech policy space, will find this essential, not just interesting. I can absolutely see why this is a foundational text in the āAI harms / big data ethicsā conversation, but for my personal taste and attention span, it turned into more homework than compelling nonfiction.
Would I recommend it? If you're already fluent in data science, tech ethics, or policy, absolutely, non-negotiable, add it to your shelf immediately. Iād recommend Weapons of Math Destruction to readers who are comfortable with more technical explanations and want a foundational, early look at how algorithms scale inequality. For me, this landed as a respectful DNF at 40%.