The first incentive for the project arose from the current controversial development in the world of media content as well as a fascination for machine learning and data analytics. The project examines the behaviour of users on the internet and the ever growing dangers of targeted, politically motivated misinformation in modern media.

The resulting website showcases canards (which are generated weekly) in three selectable designs and records a wide range of user data (selected design, referral site, the users location) by sending it to a self-hosted analytics platform. The synthesis of these articles is managed by an algorithm that combines the most recent and advanced methods of natural language processing (including, but not limited to “Syntactic Dependency Analysis”, “Named Entity Recognition”, “Word Morphology Analysis”) and public knowledge bases.