This randomized controlled trial examined the effectiveness of an online autism training intervention in reducing stigma toward autistic individuals. Participants were 208 Korean undergraduate students who were blinded to group allocation, with 106 assigned to the autism training and 102 assigned to a control intervention. All participants completed an online Qualtrics survey that included a pre-test survey (perceived similarity to different minority groups), the training, and a post-test survey (perceived similarity, stigma toward autistic people, knowledge about autism, confidence in their knowledge, and open-responses question asking for descriptions of autism). We conducted independent sample t tests and a mixed-effects model to examine group-level differences, and a reliable change index (RCI) analysis to examine individual changes in the perceived similarity score. The responses to the open-ended questions were analyzed using qualitative content analysis. Results revealed that the experimental group reported reduced stigma toward autistic individuals, increased perceived similarity with autistic individuals, and increased confidence in their autism knowledge and demonstrated increased knowledge about autism compared to the control group. At the individual level, the RCI analysis indicated that the training was effective for a notable subset of participants (25%). At post-test, the experimental group recognized the strengths of autistic people and described autistic individuals as “people like us,” while many in the control group mentioned misconceptions about autism. Future research should employ more robust intervention designs and tailor materials for diverse target populations, including administrators, to foster systemic destigmatization.