Traffic networks are vulnerable to disinformation attacks

Abstract Disinformation continues to raise concerns due its increasing threat society. Nevertheless, a of disinformation-based attack on critical infrastructure is often overlooked. Here, we consider urban traffic networks and focus fake information that manipulates drivers’ decisions create congestion at city scale. Specifically, two complementary scenarios, one where drivers are persuaded move towards given location, another they away from it. We study the optimization problem faced by adversary when choosing which streets target maximize disruption. prove finding an optimal solution computationally intractable, implying has no choice but settle for suboptimal heuristics. analyze such heuristic, compare cases targets spread across Chicago vs. concentrated in business district. Surprisingly, latter results more far-reaching disruption, with impact felt as far 2 km closest target. Our findings demonstrate vulnerabilities may arise not only hardware software, also behavioral manipulation.