Over the years, our column series in TechTrends has explored the intersections of creativity, education, and emerging technologies. Over the past year, we’ve examined GenAI’s impact through interviews with leading thinkers, practitioner studies, and conceptual pieces on its history, promise, and risks. Our goal has not been to champion or dismiss these technologies, but to examine their potential and limitations with equal rigor.

In this latest article, co-authored with Danah Henriksen, we turn to a question that lies at the heart of education: curiosity. The common claim is that GenAI will revolutionize learning by meeting each learner’s curiosity with instant, tailored responses. But we question the assumption beneath that claim: that making it easier to satisfy curiosity will necessarily deepen learning.

Drawing on Jordan Litman’s distinction between discovery curiosity (the pleasure of exploration) and deprivation curiosity (the itch to close knowledge gaps), we argue that GenAI’s sycophantic design (its tendency to please, affirm, and provide smooth closure) may systematically feed the wrong kind of curiosity. The result is what we call the deprivation curiosity trap: a self-reinforcing cycle where the pleasure of quick answers replaces the productive discomfort of genuine inquiry.

The article is also a tribute to educational psychologist David Berliner, who passed away recently and who reminded us that education is “the hardest science of all.” His insight becomes even more urgent as we invite AI into learning spaces: the challenge is not primarily technological but psychological, and psychological challenges cannot be solved with better software alone.

You can find the complete citation and link to the article below.

Mishra, P. & Henriksen, D. (2025). The Curiosity Paradox: How Sycophantic GenAI May Undermine Learning. TechTrends, 69, 1127–1133. https://doi.org/10.1007/s11528-025-01156-z