The Adaptive Information Disclosure (AID) project is part of a larger effort that aims at the creation of a so-called Virtual Lab environment for e-science . In the context of AID, we are building a suite of dynamic, model-driven information and knowledge extraction tools on top of an architecture for grid-based distributed data analysis with VL-e funding from 2004 until the end of 2009. Continued funding for 2010 and beyond has already been pledged. Keywords are: semantic models, service-oriented architecture, Semantic Web, text mining, grid mining, grammar induction, information extraction, question answering, and the dynamic maintenance of ontologies.