Evidence from rodent studies suggests that central delivery of insulin may regulate peripheral glucose metabolism through indirect cross-talk between brain and peripheral tissues. In the present study, we aimed to investigate the acute effects of intranasal insulin on endogenous (hepatic) glucose production (EGP) and hepatic energy metabolism in humans. Eight lean normoglycemic humans (age 26 ± 2yrs, BMI 23 ± 1 kg/m2, CON) and 8 patients with type 2 diabetes (age 61 ± 3yrs, BMI 27 ± 1 kg/m2, T2D) received intranasal insulin (160 IU; Actrapid, Novo-Nordisk, Denmark) or placebo in a randomized, single-blinded, cross-over trail. Rates of EGP were assessed by the isotopic dilution technique with [6,6-2H2]glucose and metabolic parameters were monitored with repeated blood sampling. Hepatic fat content (HCL) and adenosine triphosphate (ATP) concentrations were determined at baseline and three hours after intervention by magnetic resonance spectroscopy (31P/1H-MRS) on a 3T clinical MR scanner (Philips, Best, The Netherlands). Intranasal insulin caused a transient increase in serum insulin levels at 10 minutes in CON (Δ= 3.6 ± 0.8µU/ml, p = 0.01 vs. placebo). Blood glucose concentrations were not affected by either intranasal insulin or placebo and remained unchanged in both CON and T2D. The hepatic insulin sensitivity index, determined from EGP and insulin levels, was not changed after intranasal insulin in either group compared to placebo, but was lower in T2D for 180 min (9.7 ± 0.9 vs. 13.5 ± 0.7, p = 0.002). HCL decreased (Δ= 0.2 ± 0.1%, p = 0.04), whereas absolute concentrations of hepatic γ-ATP increased (Δ= 0.5 ± 0.2 mmol/l, p = 0.04) after intranasal insulin in CON, but not in T2D. In conclusion, central insulin has no acute effect on glucose production in humans, but associates with beneficial effects on hepatic lipid and energy metabolism, both of which are blunted in T2D. This could either result from greater insulinemia and/or higher insulin sensitivity in the normoglycemic humans.