Mobile computing and agent technologies are receiving a great deal of interest as they have a lot to offer towards achieving the vision of usable distributed systems in a heterogenous network environment such as the Internet. The ability to move computations across the nodes of a wide area network helps to achieve deployment of services and applications in a more flexible, dynamic and customisable way than the traditional client-server distribution paradigm. However fundamental challenges remain in the area of security. In this talk, we will address some of the key issues in secure networked computing. We will first address the developments that have taken place in distributed computing which as led to greater mobility and networked applications in both local area and wide area networks. Then we will consider some of the characteristics of network based computing and in particular the development of mobile agent based applications. We will then briefly address the notion of trust and describe how mobile agents violate some of the common trust assumptions made in the design of secure systems. Then we will outline a security model for mobile agents and the provision of security services for agent based applications. We will conclude by identifying some of the open security issues that need to be investigated.