Aligning with Human Values to Enhance Interaction: An eHMI-Mediated Lane-Changing Negotiation Strategy Using Bayesian Inference

As autonomous driving technology evolves, ensuring the stability and safety of Autonomous Driving Systems (ADS) through alignment with human values becomes increasingly crucial. While existing research emphasizes the adherence of AI to honest ethical principles, it overlooks the potential benefits of benevolent deception, which maximize overall payoffs. This study proposes a game-theoretic model for lane-changing scenarios, incorporating Bayesian inference to capture dynamic changes in human trust during interactions under external Human-Machine Interface (eHMI) disclosed information. Case studies reveal that benevolent deception can enhance the efficiency of interaction in up to 59.4% of scenarios and improve safety in up to 52.7%. However, in the most pronounced cases, deception also led to trust collapse in up to 36.9% of drivers, exposing a critical vulnerability in the ethical design of ADS. The findings suggest that aligning ADS with comprehensive human ethical values, including the conditional use of benevolent deception, can enhance human-machine interaction. Additionally, the risk of trust collapse remains a major ethical loophole that must be addressed in future ADS development.