Networked virtual environments (NVE) refer to the category of distributed applications that allow a large number of distributed users to interact with one or more central servers in a virtual environment setting. Recent studies identify that malicious users may compromise the semantic integrity of NVE applications and violate the semantic rules of the virtual environments without being detected. In this paper, we propose an efficient audit protocol to detect violations of semantic integrity through a probabilistic checking mechanism done by a third-party audit server. Key words: networked virtual environments, algorithm, audit, integrity