> Could you expand on what you mean by the 'abuse of knowledge'?

A nice way of saying "lying". Learning how to game the statistics (e.g. publish the 20th experiment that showed significance but fail to mention the other 19).