In this paper, our objective is on the study and analysis of Sensitive Information in Relational Database. In recent days we have seen that various information or data is hack by the unauthenticated person. So based upon that concept our main approach is to provide secrecy and privacy. For that purpose we characterize sensitive data or information as the extensions of secrecy views. The database, before returning the answers to a query fired by any restricted user, is updated to make the secrecy views empty or a single tuple showing only null values. Then, a query about any of those views returns no meaningful information because of these the database is not physically change but whatever updates are done is only virtual and minimal. Minimality makes sure that query answers, while being privacy preserving, are also maximally informative. Whatever virtual updates are proportional to the null values as used in the SQL standard. We provide the semantics of secrecy views, virtual updation, and secret answers to queries.