The algorithm knows what you like. Too well. I examined how Netflix's recommendation engine systematically narrows viewer choices, creating intellectual bubbles that reinforce existing preferences while quietly eliminating exposure to challenging or unfamiliar content. What does it mean when the world's largest streaming platform optimizes for comfort over discovery?
Read on Medium →Everyone talks about AI bias in the abstract. I went looking for concrete evidence. This piece documents my investigation into real-world machine learning systems, the patterns of discrimination I uncovered, and why the industry's standard responses are inadequate. The problem isn't that AI is biased—it's that we've built systems where bias is a feature, not a bug.
Read on Medium →Data tells stories, but sometimes those stories are stranger than fiction. This exploration dives into the peculiar findings that emerge when you look closely at datasets—from acoustic emissions in stressed plants to statistical anomalies that challenge our assumptions about the natural world. A reminder that the best insights often hide in the data points everyone else ignores.
Read on Medium →