This episode answers a question every leader deploying AI needs to ask: why do two people get wildly different value from the exact same AI model?
A new MIT Sloan and Stanford study asked 1,000 adults to write their own prompts asking GPT-5.2 and Gemini 3 Flash for financial advice, then simulated a lifetime of following it. The advice was strong across the board, but people with sharper financial literacy wrote sharper prompts, and that gap alone was worth roughly five percent more wealth at retirement.
Soren breaks down why the researchers call this a "demand" problem, not a model problem, and why smarter AI won't close the gap on its own, it will widen it. He connects the finding to what's happening inside companies right now, where employees who frame clear, detailed prompts get measurably better output than those who don't, and argues that asking a sharp question has become a core business skill, not just a personal finance one.
Leaders walk away with three moves: teach people to frame requests with real numbers and goals before they need the answer, practice the asking itself on a live example rather than the topic in the abstract, and tie every lesson to a real decision, like a job offer or a budget call, so it sticks.
Subscribe and listen on:
YouTube: https://www.youtube.com/playlist?list=PLU3e5k8HJeOGEa7AGA97kvVjtpi6uqG0t
Spotify: https://open.spotify.com/show/033iseLQje3RnPGqUlh2I8?si=cb19276a3c68459d
Apple Podcasts: https://podcasts.apple.com/us/podcast/leapfrogging-the-headlines-with-soren-kaplan/id1896880297
View original on Inc. Magazine: MIT and Stanford Found a 5 Percent Retirement Gap. It Comes Down to Asking Better Money Questions