Abstract. Distributed authorization provides the ability to control access to resources spread over the Internet. Typical authorization systems consider a range of security information like user identities, role identities or even temporal, spatial and contextual information associated with the access requestor. However, the ability to include computing platform related information has been quite limited due to constraints in identification and validation of platforms when distributed. Trusted computing is an exciting technology that can provide new ways to bridge this gap. In this paper, we provide the first steps necessary to achieving distributed authorization using trusted computing platforms. We introduce the notion of a Property Manifest that can be used in the specification of authorization policies. We provide an overview of our authorization architecture, its components and functions. We then illustrate the applicability of our system by implementing it in a Web service oriented architecture.