Dealing with context seems to be a major challenge which the field of Artificial Intelligence (AI) will have to face in the next few years. The importance of the challenge is borne out by the various scientific events focusing on context which have been held since 1995. This does not mean that context has not been studied before - it has in such domains as Natural Language Processing - but it has yet to be studied in depth since previous research has only covered few aspects of the problem. We present in this paper a survey of the literature dealing directly and explicitly with context whatever the domain. This makes it possible for us to have a clear view of context in AI. One of the conclusions of this survey is to point out the existence of different types of context in areas such as the representation of knowledge in a computer system, the reasoning that the system carries out using the knowledge, and the interaction the system has with people.