During the pandemic, many students faced significant challenges due to the abrupt transition to online learning and the lack of direct interaction with each other and with teachers. Therefore, it is relevant to study the contribution of conversational agents or chatbots in university education. This article aims to identify and describe the acceptance of the use of the chatbot and its impact on learning in university education, in the context of COVID-19. A systematic review was employed using a mixed-method approach with of the aim to explore the reasons behind the use of chatbots in university education to solve problems related to the lack of student interaction, motivation, as well as the lack of teaching tools and strategies. It was also identified that the criteria used to evaluate the acceptance of the use of a chatbot are design, interface and responsiveness. Finally, chatbots generated a positive impact on learning, evidenced not only in the improvement of academic performance but also in the interaction between students and teachers. The study concluded that the acceptance of the use of a chatbot and the positive impact generated on learning in university education during the context of COVID-19, will, in the current post-pandemic scenario, lead to the implementation of artificial intelligence tools such as the chatbot to traditional educational environments, transforming the way in which teaching and learning are taught, providing new opportunities for educational growth, under a more flexible and personalised learning.