AI is supposed to feel magical. So why does it so often leave product teams confused, overpromised, or quietly disappointed?In this episode, Karl sits down with Dina Atia, a PM working on AI products at Microsoft, to talk about what it is actually like to build in the middle of the hype. Dina shares candid stories from the trenches, like watching teams chase “cool” AI ideas that never quite map to real human needs, and the moment she realized AI was not a mysterious genius but also definitely not “just matrix math,” either.Along the way, she tells a surprisingly relatable story about learning to solve a Rubik’s Cube, and why that experience changed how she thinks about intelligence, perception, and expectation‑setting with stakeholders. They also dig into the emotional side of AI adoption, including the fear of being made redundant, the discomfort around privacy and environmental impact, and the quiet tension PMs feel when bold visions collide with very incremental progress.This is a grounded, honest conversation about hype, disillusionment, and what it takes to keep building thoughtfully when everyone else wants miracles. If you have ever felt caught between excitement and skepticism, or responsible for translating big promises into something real, this one will hit close to home.Productly Speaking: where product management gets real.Key Quotes: “Not understanding the problem is a huge one… especially in the AI space.” “AI isn’t thinking. It’s predicting the next word.” “Bring back engineers being lazy! What is the minimum we can do to solve this problem?” “Nobody achieves anything significant alone.” Resources & References: Gartner Hype Cycle: https://www.gartner.com/en/research/methodologies/gartner-hype-cycle  Freytag’s Pyramid: https://writers.com/freytags-pyramid  Tools mentioned:  GitHub Copilot: https://github.com/features/copilot  Lovable: https://lovable.dev/  V0: https://v0.app/  Cursor: https://cursor.com/  Connect with Dina Atia: https://www.linkedin.com/in/dinaatia/