The etiology of autism spectrum disorder (ASD) is not fully understood. There is concern about the potential association between prenatal antiseizure medication (ASM) exposure and the development of ASD. This study examined the risk of ASD in children exposed to ASMs during pregnancy. We conducted a retrospective population-based cohort study using data from the Manitoba Center for Health Policy (MCHP), which included data from pregnancies in Manitoba from January 1, 1998, to March 31, 2021. We included all live singleton births and excluded multiple births and stillbirths. We reported crude and adjusted hazard ratios (aHRs) and 95% CIs using Cox regression hazard model. To account for familial confounding, we created a sub-cohort of randomly selected one child per mother. We examined associations in pregnant women with epilepsy to address confounding by indication. Among the 289,794 pregnancies included, 2,474 (0.9%) were exposed to ASMs during the second and third trimesters of the pregnancy, with 99 (4%) diagnosed with ASD. Among 1,853 children of pregnant women with epilepsy, 755 (40.7%) were exposed to ASMs, with 30 (4%) diagnosed with ASD. The mean follow-up periods were 11.1 and 11.3 years, respectively. Primary analysis showed ASM exposure was associated with an increased risk of ASD (aHR, 1.27; 95% CI, 1.03–1.56), with lamotrigine showing a twofold increase (aHR, 2.16; 95% CI, 1.16-4.02). In the epilepsy cohort, ASM exposure doubled the risk of ASD (aHR, 2.12; 95% CI, 1.09–4.11), with phenytoin showing a nearly threefold increase (aHR, 2.68; 95% CI, 1.05–6.73). Sensitivity analysis adjusting for familial confounding showed adjusted hazard ratios of 1.13 (95% CI, 0.85–1.51) for all pregnant women and 1.50 (95% CI, 0.61-3.67) for those with epilepsy. We observed an increased risk in the overall analysis of antiseizure medication use during pregnancy and the risk of ASD in newborns. This association was also observed within the epilepsy cohort. This observed risk could either be a true effect of these medications, attributed to the correlation between siblings, confounding by severity, or an effect of residual confounding. Additional studies are warranted to identify the true risk of these medications on ASD.