A system and method for multi-level security filtering of model representations. Elements of an architecture and systems engineering model have associated security characteristics identifying the required level of security that a user must have in order to be able to obtain access to information about that element of the model. Based on these security characteristics and the security level of a user attempting to access the model, the content of the model is modified such that elements that are not accessible by a user having that security level are obscured or not visible to the user. The other elements of the model which are accessible by the user based on his/her security level are modified in the modified model so as to not provide information that would disclose the nature, character, or in some cases even the presence, of the elements that are not accessible by the user based on his/her security level.