Medical errors are a troubling issue and physicians should be careful to scrutinize their own decisions, remaining open to the possibility that they may be wrong. Even so, doctors may still be overconfident. A survey was here conducted to test how medical experience and self-confidence can affect physicians working in the specific clinical area. Potential participants were contacted through personalized emails and invited to contribute to the survey. The “risk-intelligence” test consists of 50 statements about general knowledge in which participants were asked to indicate how likely they thought that each statement was true or false. The risk-intelligence quotient (RQ), a measure of self-confidence, varies between 0 and 100. The higher the RQ score, the better the confidence in personal knowledge. To allow for a representation of 1000 physicians, the sample size was calculated as 278 respondents. A total of 1334 individual emails were sent to reach 278 respondents. A control group of 198 medical students were also invited, of them, 54 responded to the survey. The mean RQ (SD)of physicians was 61.1 (11.4) and that of students was 52.6 (9.9). Assuming age as indicator of knowledge, it was observed that physicians ≤34 years had a mean RQ of 59.1 (10.1); those of 35–42 years had 61.0 (11.0); in those of 43–51 years increased to 62.9 (12.2); reached a plateau of 63.0 (11.5) between 52–59 years and decreased to 59.6 (12.1) in respondents ≥60 years (r2:0.992). Doctors overestimate smaller probabilities and under-estimate higher probabilities. Specialists in gastroenterology and hepato-biliary diseases suffer from some degree of self-confidence bias, potentially leading to medical errors. Approaches aimed at ameliorating the self-judgment should be promoted more widely in medical education.