Parallel Data Mining (PDM) is currently attracting much research. Objects involved with ‘Parallel Data Mining' include special types of entities with the ability to migrate from one processor to another where it can resume / initiate its execution. In this article we consider security issues that need to be addressed before these systems in general, and ‘parallelized systems' in particular, can be a viable solutions for a broad range of commercial tools.