Iodine deficiency (ID) poses a significant global public health challenge. This study aimed to analyze trends from 1990 to 2021 and project future patterns ≤2050 using the extreme gradient boosting (XGBoost) model, with Shapley additive explanations (SHAP), to identify key factors and inform public health strategies. Data on ID from the Global Burden of Disease 2021 study were used to model and predict its burden ≤2050 using XGBoost, with SHAP enhancing model interpretability. In 1990, global incident cases of ID were 7.51 million (age-standardized incidence rate [ASIR]: 126.11/100,000), rising to 8.08 million by 2021 (ASIR: 105.99/100,000, a 15.96% decrease), and projected to reach 8.48 million by 2050 (ASIR: 108.20/100,000). Prevalent cases increased from 146.42 million in 1990 (age-standardized prevalence rate [ASPR]: 2801.80/100,000) to 180.81 million in 2021 (ASPR: 2213.98/100,000, a 20.98% decrease), with 194.51 million expected by 2050 (ASPR: 1900.01/100,000). Disability-adjusted life years dropped from 2.46 million in 1990 (age-standardized disability-adjusted life year rate [ASDR]: 46.19/100,000) to 2.25 million in 2021 (ASDR: 27.67/100,000, a 40.10% decrease) but are projected to rise slightly to 2.51 million by 2050 (ASDR: 25.51/100,000). SHAP analysis identified iodized salt coverage as a key factor, with higher coverage levels associated with reduced ID burden in most countries. Women and people aged 10-30 y had higher incidence rates, although prevalence and disability-adjusted life years peaked among those aged 20-45 y. Central and Eastern Sub-Saharan Africa and South Asia will continue to bear the highest burden through 2050. The XGBoost+SHAP model identified age, sex, and iodized salt coverage as key factors, with women and younger populations being high-risk groups. Strengthening iodization programs, improving health care access, targeted education, and consistent monitoring of vulnerable populations are essential to mitigate future risks and improve health outcomes.