AutoDetect is a research project funded by the State-Foundation Baden-Württemberg in which we investigate misbehavior detection mechanisms in automotive systems. The goal of our research is the development of approaches and mechanisms which can detect data and sensor values injected by attackers in internal vehicular networks and V2X communication by different detectors. Detectors can be based on consistency and plausibility checks and used to rate, filter, or even correct data. To this end, we develop a framework based on subjective logic.