When people are confronted with research that contradicts their own personal experiences, they tend to deny the science. Using a secondary multinational data set collected during the early stages of the coronavirus disease 2019 (COVID‐19) pandemic (N = 46,490), we tested this “seeing is believing” effect as it relates to the link between direct and indirect personal experience with COVID‐19 and public health behaviors (PHBs) through COVID‐19 conspiracy beliefs and perceived risk of infection. Indirect experience with COVID‐19 was associated with lower endorsement of COVID‐19 conspiracy beliefs, which negatively predicted risk perception of infection, and, in turn, positively predicted PHBs. However, direct experience positively predicted COVID‐19 conspiracy beliefs, while it negatively predicted perceived risk. Moreover, while indirect experience positively predicted PHBs, direct experience largely negatively predicted PHBs. Implications of these findings as it relates to the “seeing is believing” effect are discussed.