Paths of Curiosity in an AI World
Q3 - AI in Data, Behavior, and Human Systems with Berkeley Dietvorst, BS, PhD
Why Seeing an Algorithm Fail Makes You Trust It Less — Even When It's Still Better Than You Paths of Curiosity in an AI World What if the algorithm was actually right and you still refused to trust it? Dr. Berkeley Dietvorst discovered that people abandon algorithms the moment they see them make a mistake — even when those algorithms consistently outperform humans. And then he found something even more surprising: giving people just a tiny amount of control over the output is often enough to fix it. In this episode, we sit down with Dr. Dietvorst, Professor of Marketing at the University of Chicago Booth School of Business and the researcher who coined the term algorithm aversion. Together, we explore: What algorithm aversion actually is — and why seeing an algorithm perform makes people less likely to use it, not more Why a small ability to tweak an algorithm's output changes everything psychologically The critical difference between a prediction and a decision — and why algorithms should only ever be responsible for one of them What randomness in a domain has to do with whether you should trust an algorithm at all Why hiring, parole, and medical diagnosis feel different — and what your resistance to algorithms there might actually be telling you What a student sorted by algorithms their entire life should understand about the systems shaping their choices Dr. Dietvorst also shares his own unexpected path — from a finance major fascinated by investing to a behavioral scientist studying why humans make the decisions they make — and why following what genuinely interested him at each step was the only strategy that worked. This conversation will change how you think about every algorithm nudging your decisions, from college applications to career choices to what your feed decides you should see next. Because in a world where algorithms are often better than us at prediction, the real question was never whether to trust them. It's understanding exactly what they can and cannot do — and who gets to make the final call. LinkedIn Post: What if the algorithm was right and you still refused to use it? Dr. Berkeley Dietvorst discovered that people abandon algorithms the moment they see them make a mistake — even when those algorithms consistently outperform humans. He coined a term for it: algorithm aversion. And then he found something even more surprising: giving people just a tiny amount of control over the output is often enough to fix it entirely. In our latest episode of Paths of Curiosity in an AI World, we sat down with Dr. Dietvorst, Professor of Marketing at the University of Chicago Booth School of Business. Here's the framing that changed how I think about every AI tool I use: Algorithms are for predictions. Humans are for decisions. A prediction tells you there's a 20% chance of rain. The decision — whether to go to the outdoor concert anyway — depends on how much you hate getting wet, how much you love the band, and how you'd feel if you stayed home and it didn't rain. No algorithm can know that. Only you can. The same logic applies to college admissions, hiring, parole, medical diagnosis. The resistance people feel toward algorithms in those domains isn't always aversion. Sometimes it's the right instinct — because those aren't just prediction problems. They're value problems. And value problems belong to humans. What else we got into: Why seeing an algorithm perform makes people trust it less, not more How a tiny tweak to the output changes everything psychologically Why some domains are too random for any algorithm to ever look good — even when it's still beating humans What growing up sorted by algorithms does to how young people make decisions Dr. Dietvorst started as a finance major drawn to investing, took one class on human decision-making, and never looked back. His advice: take the econ class, take the psychology class, take the statistics class — and then go read the actual academic papers, not the summaries. That's where the real skill starts to build. When was the last time you trusted an algorithm's prediction but made a different decision anyway — and were you right to? #AI #BehavioralScience #AlgorithmAversion #DecisionMaking #FutureOfWork #PathsOfCuriosity Paths of Curiosity in an AI World Exploring tomorrow’s careers through the lens of AI Follow us: @pathsofcuriosity Website: https://pathsofcuriosity.com/ Youtube: youtube.com/@PathsofCuriosity X: https://x.com/PathsCuriosity LinkedIn: https://www.linkedin.com/in/paths-of-curiosity-in-an-ai-world-2117803aa/
