412- Nuclear Bombs, F-35s, and Herding Cats w/Stephen Salaka

Stephen Salaka has a PhD in psychology and builds nuclear weapons systems. Then he learned the hardest truth in tech: AI can't fix what humans won't adopt. Stephen Salaka wanted to build nuclear bombs. A lab accident sent him to computer science. A stint in Japan taught him his real superpower: making humans actually use the technology he builds. Now he's a CTO with a PhD in Industrial/Organizational Psychology. That combination makes him dangerous to every AI myth floating around C-suites. "The biggest fundamental misconception is AI is going to be a panacea for everything," he says. "The real trouble most organizations face is the people." We get into why UPS's maintenance system failed for years until Stephen added change management, how vibe coding should stop at the prototype stage, and why the AI bubble collapse is coming faster than anyone thinks. Plus his framework for bringing order to chaos without mandates. The payoff: Stephen's lived at the fault line between brilliant technology and stubborn humans. He knows which one wins. Key takeaways: Change management isn't optional - it's the actual product you're delivering; Vibe coding works for prototypes, then convert to proper backend components; Smooth processes first, then speed follows - not the other way around