Background: Attempts to conduct meta-analyses of the association between child development and diabetes have been limited by the wide range of tools and definitions of developmental outcomes used in the literature. We aim to meta-analyze a widely used measure of child development, the Bayley Scales of Infant Development, with respect to exposure to diabetes and developmental scores. Methods: PsycINFO, MEDLINE/PubMed, EMBASE, Emcare, and Google Scholar databases were searched. Two independent reviewers screened, extracted, and quality-appraised the studies using JBI SUMARI software. Forest plots were created with the standardized mean difference using the random-effects model, and heterogeneity was assessed using I2. Results: Seven studies were identified. The pooled results on psychomotor and mental development index mean scores were lower for infants born to mothers with diabetes than for the control group (Cohen’s d = −4.49, df = 7, I2 = 0%, p = 0.001 and Cohen’s d = −3.4, df = 9, I2 = 27%, p = 0.001, respectively). Effects were larger in infants born to mothers with type 1 and 2 diabetes and at age 12 months. Conclusions: Maternal diabetes should be considered as a risk factor for children’s development, mainly when born to mothers with pre-existing diabetes.