Knowledge Management: an AI approach Knowledge Management covers a wide range of issues and different solutions have been offered by people from a variety of disciplines. KM initially was stemmed from the business management community recognising the fact that the advance of modern technologies have brought lots of information to people, but not knowledge. However, in order to make the right decisions and do the right things at the right time, knowledge is needed, not just information. Knowledge Management covers issues such as business practices, social environment and information science. And, indeed scholars and practitioners of different disciplines have offered specialised technologies, methodologies and empirical methods to assist the process of KM. What is interesting is the fact that it is increasingly recognised that automatic support can help to structure and speed up the management of knowledge. Furthermore, it is no longer satisfactory with merely the manipulation on the syntactic level, but the manipulation must also operate at the semantic level. All of these provide AI a great opportunity to contribute to KM. In this talk, I will briefly introduce KM and some existing KM approaches, followed by approaches that AI may contribute or have contributed to KM. I would like to propose a Multi-Perspective Knowledge Modelling Framework which provides a structure way to capture the initially informal and possibly diverse input in a more coherent way and to suggest how formal methods may be used in this framework to assist the process of KM.