Dental caries is a prevalent chronic disease influenced by both genetic and environmental factors, particularly dietary. Using a 2-sample Mendelian randomization (MR) approach with genome-wide association study data, this study examined the potential causal relationship between sugar preference, diabetes, and dental caries. The research seeks to clarify the influence of genetic predispositions and metabolic conditions, such as diabetes, on dental health. We conducted a 2-sample MR analysis using genome-wide significant single nucleotide polymorphisms (P < 5 × 10⁻⁸), linkage disequilibrium pruning (R2 < 0.001), and multiple MR methods including inverse-variance weighted, MR-Egger, robust adjusted profile score, and MR-PRESSO. Genetic variants associated with exposures were harmonized with outcome data using the 2 sample MR R package, and ambiguous palindromic single nucleotide polymorphisms were excluded. We found no strong evidence for a causal effect of genetically predicted sugar intake on dental caries. However, both type 1 and type 2 diabetes showed positive associations with caries experience, suggesting that metabolic conditions may contribute to caries risk independently of sugar intake. No significant genetic correlation was observed between sugar preference and diabetes, and Steiger filtering supported the directionality of our analysis. Findings from sensitivity analyses (e.g., leave-one-out, MR-PRESSO) confirmed the robustness of the results. Our findings suggest that the traditionally assumed causal role of sugar intake in dental caries may be overstated. Instead, they support a model in which genetic predispositions and metabolic disorders such as diabetes may play a more substantial role. These findings advocate for personalized dental care approaches that consider individual genetic predispositions and metabolic conditions, offering new insights for future research and clinical practice.