Abstract: Various technologies described herein relate to routing an autonomous vehicle based upon likelihood of interacting with a disruptive third-party vehicle. A computing system receives an origin location of an autonomous vehicle and a destination location of the autonomous vehicle. The computing system identifies a route for the autonomous vehicle to take based upon output of a computer-implemented model. The computer-implemented model is generated based upon labeled data that is indicative of observed instances of a disruptive third-party vehicle exhibiting behavior that impacts operation of an autonomous vehicle. The output of the computer-implemented model is a score that is indicative of a likelihood that the disruptive third-party vehicle will be traveling through a geographic location along a candidate route at a certain time. The score is used in part to identify the route from amongst a plurality of candidate routes. The autonomous vehicle then travels along the route.