In recent years, computational toxicology as a discipline has received considerable attention as it offers the attraction of being able to provide rapid assessments of toxicological hazards and/or risks at a significantly lower cost than in vivo or even in vitro assays. It also has the added attraction of reducing, refining, or perhaps even eliminating the use of animals in toxicology testing. However, just how good are the computational tools available and how should they be used to draw appropriate conclusions about chemical safety? This presentation focuses on the current state of the art and highlights some of the limitations of the current approaches in computational toxicology.