As Automated Vehicles (AVs) become more advanced, they must interact seamlessly with vulnerable road users—especially pedestrians. But how do pedestrians know when an AV has seen them? Should AVs mimic human driving cues or introduce new forms of communication? In this talk, we’ll explore the critical role of both explicit cues (e.g. external HMI) and implicit cues (e.g. AV movement and kinematics) in shaping pedestrian trust, understanding, and behaviour. Drawing on findings from over ten experimental studies in our pedestrian lab at the University of Leeds, we’ll share key behavioural insights and design implications. Whether you’re building AVs or designing their interfaces, this presentation will challenge assumptions and provide evidence-based recommendations to ensure safer, more intuitive pedestrian-AV interactions. If AVs are to succeed in complex urban environments, we must start designing with people in mind—not just systems.