Artificial intelligence (AI) is used by logistics companies to optimize delivery driver route characteristics. When creating routes, the AI makes decisions that optimize efficiency and cost reduction. These decisions, however, can significantly increase the risk of driving accidents by affecting driver stress, fatigue, attention, situation awareness (SA), and risky driving behavior. Driving accidents during work have risen by 40% from 2020 to 2021 and represent the most fatal of all workplace accidents, highlighting the importance of accident prevention. The current chapter proposes a continuous multi-method assessment of delivery drivers’ physiological, perceptual, and behavioral states throughout the work shift. Our methodology was then applied to a series of pilot tests in which a delivery driver was monitored for approximately 12 hours. The results indicated that our methodology is viable for evaluating driver fatigue, stress, attention, situation awareness, and risky driving behaviors over the course of an entire work shift. We believe that our methodology can be successfully applied in delivery contexts to reduce the rising number of driving accidents. Other applications of our methodology to enhance human-AI interaction are also discussed.