Autistic spectrum disorders (ASD) are a group of neurodevelopmental disorders that manifest as deficits in social communication and interaction, and restricted, repetitive behaviors and interests. ASD affect at least 1% of the population and are associated with lifelong disability and early death. There are no effective biological treatments for ASD, although non-invasive neuromodulation has sparked great interest as a possibly useful therapeutic approach. Here, we present preliminary results of a systematic review on the effectiveness of transcranial magnetic stimulation (TMS) in ASD treatment. Using appropriate syntax we searched Pubmed, Web of Science, Science Direct, and Educational Resources Information Clearinghouse. Following standard PRISMA statement (Preferred Reporting Items for Systematic Reviews and Meta-analyses) procedures, we selected 12 eligible studies, comprising four controlled and four uncontrolled trials on the effects of TMS on ASD core symptoms, and 9 controlled and three uncontrolled trials on TMS effects on cognitive performance in ASD. The 12 studies totaled 233 subjects. Although combined effect sizes favor TMS in all fours groups of studies, conclusions are limited by the high study heterogeneity. Furthermore, only three of the controlled studies used sham TMS as the control intervention, and only two studies followed up the therapeutic effects after the last TMS session. Side effects, none of them serious, occurred in 6.4% of treated subjects. Our main conclusion is that there is currently little evidence that sustains the commercial offer of TMS for treating ASD. Better-designed studies are badly needed to fully elucidate the role of TMS in the treatment of ASD.