Elevated neonatal birth size is associated with risk for the development of overweight and obesity in childhood. A large body of evidence has demonstrated the importance of the in utero environment in determining growth in postnatal life. Numerous studies have suggested that reduction in maternal insulin sensitivity in the maternal environment contributes significantly to fetal growth. There is an inverse relationship between birth size and both very low and very high maternal insulin sensitivity. Accordingly, marked reduction in maternal insulin sensitivity in late stages of pregnancy has been linked to the preservation of fetal nutrient availability for growth, and thereby increased offspring birth size. It has been suggested that regular exercise may reduce offspring size by regulating the nutrient supply to the fetus through increases in maternal insulin sensitivity. Few data exist on the metabolic consequences of maternal exercising during pregnancy for mother and offspring. This community-based, randomized, controlled trial investigated the effects of aerobic exercise training during the second half of pregnancy on maternal insulin sensitivity and neonatal outcomes. The participants were 84 healthy nulliparous singleton women at a mean age of 30 years, mean body mass index (BMI) of 25.5, and at less than 20 weeks' gestation. Of the 84 randomized subjects, 47 in the exercise group and 37 in the control completed both baseline and late gestation assessments. Aerobic exercise intervention was a home-based stationary cycling program performed by study subjects from 20 weeks of gestation till the delivery. The primary study outcome measures were offspring birth weight and maternal insulin sensitivity. Measures of neonatal auxology determined at birth included birth weight and BMI and were reported as standard deviation scores (SDS). Venous cord blood samples were collected at delivery to assess fetal concentrations of growth-related peptides (IGF-1 and IGF-2). The study was performed according to intention-to-treat analysis. After adjustment for gender and gestational age at delivery, exercisers had lower birth weight (SDS, control: 0.23 ± 0.8 vs. exercise: -0.19 ± 0.9; P = 0.03) and BMI at birth (SDS, control: 0.40 ± 0.9 vs. exercise: -0.01 ± 0.09; P = 0.04). The reduction in maternal insulin sensitivity occurring in late gestation was not correlated with either exercise or offspring size. Cord serum concentrations of serum IGF-1 (P = 0.03) and IGF-2 (P = 0.04) were lower in offspring of exercisers compared with control offspring. These data suggest that maternal exercise has an effect on decreased endocrine stimulation of fetal growth. The reduction in birth weight after exercise training is not associated with changes in maternal insulin sensitivity.