OBJECTIVE To evaluate whether the sodium–glucose cotransporter 2 inhibitor empagliflozin (EMPA) reduces liver fat content (LFC) in recent-onset and metabolically well-controlled type 2 diabetes (T2D). RESEARCH DESIGN AND METHODS Patients with T2D (n = 84) (HbA1c 6.6 ± 0.5% [49 ± 10 mmol/mol], known disease duration 39 ± 27 months) were randomly assigned to 24 weeks of treatment with 25 mg daily EMPA or placebo. The primary end point was the difference of the change in LFC as measured with magnetic resonance methods from 0 (baseline) to 24 weeks between groups. Tissue-specific insulin sensitivity (secondary outcome) was assessed by two-step clamps using an isotope dilution technique. Exploratory analysis comprised circulating surrogate markers of insulin sensitivity and liver function. Statistical comparison was done by ANCOVA adjusted for respective baseline values, age, sex, and BMI. RESULTS EMPA treatment resulted in a placebo-corrected absolute change of −1.8% (95% CI −3.4, −0.2; P = 0.02) and relative change in LFC of −22% (−36, −7; P = 0.009) from baseline to end of treatment, corresponding to a 2.3-fold greater reduction. Weight loss occurred only with EMPA (placebo-corrected change −2.5 kg [−3.7, −1.4]; P < 0.001), while no placebo-corrected change in tissue-specific insulin sensitivity was observed. EMPA treatment also led to placebo-corrected changes in uric acid (−74 mol/L [−108, −42]; P < 0.001) and high-molecular-weight adiponectin (36% [16, 60]; P < 0.001) levels from 0 to 24 weeks. CONCLUSIONS EMPA effectively reduces hepatic fat in patients with T2D with excellent glycemic control and short known disease duration. Interestingly, EMPA also decreases circulating uric acid and raises adiponectin levels despite unchanged insulin sensitivity. EMPA could therefore contribute to the early treatment of nonalcoholic fatty liver disease in T2D.