iBorderCtrl, software developed by the firm European Dynamics, aims to help border security screen non-EU nationals at EU borders (AlgorithmWatch 2019). By using “cheating biomarkers”, the system determines whether a person is lying during an “automated interview” using a virtual border guard system that tracks facial and body movements. If iBorderCtrl suspects the person to be lying, the system asks for biometric information and prompts a personal interview. iBorderCtrl’s purpose is “to reduce the subjective control and workload of human agents and to increase the objective control with automated means that are noninvasive and do not add to the time the traveler has to spend at the border” (ibid). That being said, the system still relies on a “human-in-the-loop principle” where real border security agents are monitoring the process.
Scholars argue that iBorderCtrl processes sensitive information while falling into a pseudo-scientific trap: detecting a lie requires complex psychological intuition and the current state of facial recognition technology is simply incompatible, inaccurate and fraught with bias (ibid). While algorithms are seen as objective and supremely competent, they too fall victim to bias and natural limitations.
