Throughout the past decade, artificial intelligence has developed rapidly, and the use of autonomous vehicles has gradually spread from the laboratory into the real world. Autonomous vehicles with high levels of intelligence can enhance the driving experience. However, autonomous vehicles are a long way from fully integrating into most people’s lives. Through the word-of-mouth effect, the intention to recommend can contribute to the growth of the autonomous driving market. Therefore, current research explores the mechanisms among the perceived risk of privacy safety, perceived defect, perceived behavioral control, intention to use, and intention to recommend through path analysis. Based on the results of 264 online questionnaires, the current study found that perceived risk of privacy safety, perceived defect, and perceived behavioral control could all have an effect on the intention to recommend by affecting the intention to use, and perceived risk of privacy safety can have a direct effect on the intention to recommend. In addition, there is a correlation between perceived risk of privacy safety and perceived behavior control.