Conversational skills, which are essential for effective social interactions and typically pose difficulties for individuals with autism spectrum disorder (ASD), include abilities such as initiating topics, engaging in back-and-forth dialog, and responding to conversational cues. Chatbots have been used in mental health fields, and the development of emerging technologies such as generative artificial intelligence (AI) has led to innovations in chatbot construction. This scoping review explores the technological affordances and applications of chatbots for conversational skill interventions in individuals with ASD. The study synthesizes findings from peer-reviewed research over the past two decades. Despite the potential of chatbots to provide dynamic and intelligent responses, the current literature has revealed significant limitations in chatbot construction and effectiveness. Most chatbots rely on rule-based response generation, which often leads to user frustration and a lack of motivation. Furthermore, the review identifies a gap in the application of advanced technologies such as large language models (LLMs) in chatbot development for conversation interventions. The study emphasizes the need for more personalized and adaptive chatbot designs that leverage generative AI technologies to cater to the diverse needs of individuals with ASD. Additionally, expanded data collection and the inclusion of larger sample sizes are needed to improve the accuracy and validity of chatbot models. Moreover, further evidence-based experimental research is needed to evaluate the intervention effectiveness of chatbots. This review provides comprehensive insights and recommendations for future research and development in the field, aiming to increase the effectiveness of chatbots in supporting conversational skill development in individuals with ASD.