In this paper, we describe an ongoing project to develop a collaborative safety mechanism based on mutual sharing of a driver's information (attention and performance). Most conventional driving safety approaches aim at assisting individual drivers by providing various types of safety information to the driver in the vehicle. However, the effect of such a system highly depends on the driver's response (performance), and it is unreasonable to expect sufficient performance in cases where the driver's cognitive performance is decreased due to fatigue, aging, or other factors. In our approach, we aim to share information on the driver's current performance level with other drivers and pedestrians. Using this sharing mechanism, the various people in a traffic environment will be able to better predict drivers' behavior, allowing them to take proactive actions to prevent accidents. We are now developing a system providing such a mechanism using both a performance-detection algorithm and an information-sharing method. For the detection part, we employ a behavior-based driver-evaluation method using multiple sensory devices. For information sharing, we are investigating both a method using vehicle-mounted displays and a method using wireless communication. In addition, we have also conducted a large-scale questionnaire investigation, involving more than 1,500 participants, to clarify the social acceptance of the proposed mechanisms.