Objective：To analyze the related influencing factors of hepatic steatosis and explore the relationship between the arterial sclerosis index (AI) and non-alcoholic fatty liver disease (NAFLD). Methods：This study was a secondary analysis based on the public data of a cross-sectional study conducted at the Medical Health Examination Center of Kurosawa Memorial Hospital in Japan, involving 856 participants aged 24 to 84 years. The participants were divided into the NAFLD group (n=209) and the non-NAFLD group ( n=647) according to whether they had non-alcoholic fatty liver disease. The differences in clinical general data, laboratory indicators, AI values, and the detection rate of arterial sclerosis between the two groups were compared. Logistic regression analysis was used to evaluate the related factors of NAFLD, and a regression model was constructed to assess the correlation between AI and NAFLD. The receiver operating characteristic (ROC) curve was used to test the value of AI as an indicator for risk assessment of NAFLD. Results：There was a statistically significant difference in the gender composition ratio between the NAFLD group and the non-NAFLD group (P < 0.001), with more males in the NAFLD group. Compared with the non-NAFLD group, the NAFLD group had higher levels of body mass index (BMI), systolic blood pressure (SBP), diastolic blood pressure (DBP), aspartate aminotransferase (AST), alanine aminotransferase (ALT), γ-glutamyl transferase (GGT), fasting blood glucose, uric acid, triglyceride (TG), total cholesterol (TC), low-density lipoprotein cholesterol (LDL-C), and arterial pulse wave velocity of brachial artery (baPWV) (P < 0.001), lower levels of high-density lipoprotein cholesterol (HDL-C) (P < 0.001), and less exercise habits (P = 0.001). The AI values and the detection rate of arterial sclerosis in the NAFLD group were higher than those in the non-NAFLD group (P < 0.001). Binary multivariate Logistic regression analysis showed that AI, BMI, AST, fasting blood glucose, and baPWV were positively correlated with NAFLD. After fully adjusting for confounding factors, binary multivariate Logistic regression analysis showed that AI was positively correlated with NAFLD (OR = 1.846, 95% CI=1.541 - 2.121, P < 0.001), and the population in the highest AI group showed a significantly higher tendency of NAFLD compared to the lowest group (OR = 6.169, 95% CI = 3.006 - 12.661, P < 0.001). In the generalized additive model, there is a direct linear relationship between AI and NAFLD (log-likelihood ratio test, P = 0.949). The area under the ROC curve of AI is 0.775 (95% CI=0.739 - 0.811, P < 0.001), which is higher than that of baPWV and traditional lipid indicators [baPWV (AUC = 0.616, 95% CI = 0.574 - 0.659, P < 0.001), TC (AUC = 0.576, 95% CI = 0.532 - 0.621, P = 0.001), TG (AUC = 0.735, 95% CI = 0.696 - 0.774, P < 0.001), LDL-C (AUC = 0.639, 95% CI = 0.597 - 0.681, P < 0.001)].Conclusion：AI, BMI, AST, fasting blood glucose, and baPWV are independent related factors for NAFLD. Although AI cannot directly serve as a causal indicator for NAFLD, its potential in risk assessment is worthy of attention, especially when combined with imaging examinations, which can help improve the early diagnosis rate of NAFLD.