This research investigates how Vision-Language Models (VLMs) can transform automotive safety through improved driver monitoring systems (DMS).

• Leverages zero-shot capabilities of modern VLMs to detect driver states and behaviors
• Enables more accurate identification of distraction, fatigue, and potentially dangerous situations
• Creates a paradigm shift from traditional data-intensive approaches to prompt-based solutions
• Enhances vehicle security systems with more intelligent, adaptable monitoring

This innovation matters for security because it strengthens automotive safety infrastructure through more sophisticated driver monitoring, potentially reducing accidents caused by human factors and creating more responsive vehicle security systems.

Exploration of VLMs for Driver Monitoring Systems Applications