Criminals allegedly use effective novelty to intentionally exploit and harm others (creative fraud, theft, and murder). However, empirical evidence that criminals possess higher malevolent creativity than individuals without criminal backgrounds is lacking. We compared a male sample of prisoners in a maximum-security penitentiary (n = 140), police officers (n = 122), and the general population (n = 106) on three different aspects of malevolent creativity: self-reported malevolent creativity behavior (MCBS), willingness to engage in malevolent creativity on a test (MCT), and malevolent creativity potential on that test (reduced n = 285). Group comparisons (ANOVAs) differed for different malevolent creativity aspects: Prisoners reported more malevolent creativity behavior in daily life (MCBS) than nonprisoners, which may reflect their alleged criminal personality or the effects of confinement on creative coping with threat. However, prisoners also performed worse than police officers in generating creative ideas for taking revenge on others (MCT). No differences in initial willingness to engage in malevolent creativity (MCT) emerged. This discrepancy of self-report and ability is discussed from several angles, including suitability of the applied measures and heterogeneity of prison populations. This study constitutes the first empirical insights into the often hypothesized but rarely tested malevolent creativity expression in the criminal mind.