background Exercise addiction (EA) is manifested by loss of control over exercising and experiencing numerous negative consequences. One of the most commonly used psychometric instruments to assess the risk of exercise addiction is the six-item Exercise Addiction Inventory (EAI). The present study translated and validated the EAI into Polish and evaluated its psychometric properties. participants and procedure The study included 858 adult participants (54% female) aged between 18 and 69 years (M = 35.20, SD = 9.54) who exercised recreationally. Participants completed the EAI along with exercise-specific questions and basic demographics. Additionally, a subsample (n = 214) completed the Exercise Dependence Scale-Revised. Confirmatory factor analysis (CFA) was conducted to assess the construct validity of the Polish EAI. Convergent and criterion validity, as well as reliability, were also assessed. results The CFA confirmed the one-factor model with good fit statistics (comparative fit index and Tucker-Lewis index > .95, root-mean square error of approximation < .08). The EAI was invariant between men and women. Convergent validity was adequate. The scale’s internal consistency was excellent (Cronbach’s α = .91, McDonald’s ω = .91, and composite reliability = .91). conclusions The present study successfully validated the EAI among Polish adults. The results provide robust evidence concerning the scale’s reliability, construct validity, and convergent validity, enabling further research on exercise addiction in Poland.