Evaluating Gaze-Manipulation Risks via Image Movement Attacks on Optical See-Through AR/MR HMDs

Optical see-through Augmented Reality and Mixed Reality head-mounted displays are increasingly adopted for enhanced task efficiency and user experience. However, they also introduce emerging security and privacy threats. This study investigates Image Movement Attacks, a form of Visual Perceptual Manipulation Attacks where visual elements on a head-mounted display are subtly repositioned to covertly influence gaze and head orientation, potentially leading to privacy intrusion or manipulated user awareness. This paper presents three key contributions: the proposal of realistic attack scenarios, the experimental validation of these threats, and an exploration of fundamental countermeasures. As one of the realistic attack scenario, a supply chain attack is presented, such as embedding malicious programs into assets used for application development. A user study with 22 participants using the Magic Leap 1 device demonstrated that continuous visual movements often went unnoticed, posing significant risks for covert gaze manipulation. The findings indicate that covert image movements alter gaze and head orientation, raising concerns that Image Movement Attacks could trigger serious issues in an impactful attack chain or even manipulate everyday decisions. To mitigate these risks, this study explores a certification framework for third-party assets as well as platform-level anomaly detection through monitoring and integrity checks.