R. Scott Kemp, instructor of MIT's 22.04 Social Problems of Nuclear Energy, examines why nuclear experts and ordinary citizens so often reach opposite conclusions about nuclear power's risks. The lecture, the nineteenth in the Fall 2025 course, works through the psychological and institutional roots of this gap, including how experts assess probability and consequence differently from laypeople, how trust in institutions shapes risk perception, and how past events like Chernobyl and Fukushima get processed into public memory. Kemp lays out concrete steps the nuclear industry and regulators could take to close the gap, from communication strategies to changes in how safety decisions get made and explained. It is a discussion-style lecture grounded in survey data and case history rather than engineering detail, aimed at explaining disagreement rather than resolving it by fiat.