The role of artificial intelligence in the field of migration is gaining increasing relevance worldwide, especially in the development of “smart borders”. That concept includes the use of different technologies with the scope to implement a more effective and planned management of incoming migration flows. By adopting a constitutional perspective, this paper focuses on the challenges that automated decisions at borders poses for the rights of asylum seekers in the context of the European Union. Particular attention will be paid to the control system called iBorderCtrl, recently experimented along some European borders under strong and unauthorized migration pressure. This algorithm aims at detecting potential deceptions of migrants on the basis of facial recognition technology and the measurement of micro-expressions. The range of potentialities and risks will be thus analyzed taking into account the lack of a human oversight during a “life changing” phase for the individuals involved.