An interface securing method between a user and a non certified application running on an electronic device and providing at least one view to be rendered on a screen, and a view comprising or associated to at least one interface control component. The method implements, prior to rendering a view: analysis of the interface control components of the view, so as to detect at least one interface control component that could pose a security problem, called a suspicious component; and if at least one suspicious component is detected, restriction or invalidation of the functions of at least one input control component, distinct from the at least one suspicious component and able to allow the information to be provided by the user to the application.