How do we stop from becoming so reliant on LLMs that we lose our ability to think. A new paper explores answers based on an unlikely source: rigorous mathematical models. In this episode, Cal explains the paper and its conclusions, arguing that once you cut through its complexity it yields a useful way of thinking about our current AI moment. More from Cal Download Cal’s FREE guide to cultivating a deeper life: calnewport.com/ideas Learn more about Cal’s books: calnewport.com/books Listen to Cal’s podcast: thedeeplife.com/listen Resources Mentioned: https://arxiv.org/pdf/2609.03344 https://thedispatch.com/article/artificial-intelligence-hugging-face-security-death/ Sponsors: https://www.factormeals.com/deep50off (Use code “DEEP50OFF”) https://www.calderalab.com/deep (Use code “DEEP”) https://www.notion.com/cal https://www.mybodytutor.com Credits: Podcast Production: Jesse Miller Newsletter/Research: Nate Mechler Theme Music: Jay Kerstens