Non-alcoholic fatty liver disease (NAFLD) is currently the most prevalent chronic liver disease worldwide. At the same time, the relationship between air pollution and the likelihood of developing NAFLD has been a subject of debate due to conflicting findings in previous observational research. Our objective was to examine the potential correlation between air pollutant levels and the risk of NAFLD in the European population by employing a two-sample Mendelian randomization (MR) analysis. The UK Biobank Consortium provided the summary statistics for various air pollution indicators (PM_2.5, PM_2.5 absorbance, PM_2.5–10, PM₁₀, NO₂, and NO_x). Additionally, information on NAFLD was obtained from three studies, including one derivation set and two validation sets. Heterogeneity, pleiotropy, and sensitivity analyses were performed under different MR frameworks, and instrumental variables associated with confounders (such as education, smoking, alcohol, and BMI) were detected by tools. In the derivation set, causal relationships between PM_2.5, NO₂, and NAFLD were observed in univariable Mendelian randomization (UVMR) (Odds Ratio (OR) = 1.99, 95% confidence interval (95% CI) = [1.22–3.22], p = 0.005; OR = 2.08, 95% CI = [1.27–3.40], p = 0.004, respectively). After adjustment for air pollutants or alcohol intake frequency in multivariable Mendelian randomization (MVMR), the above genetic correlations disappeared. In validation sets, the null associations remained in UVMR. Our findings from MR analysis using genetic data did not provide evidence for a causal association between air pollution and NAFLD in the European population. The associations observed in epidemiological studies could be partly attributed to confounders.