How is Voting Theory Really Useful in Multiagent Systems?

For more than a decade it has been taken for granted that social choice theory can inform the design of multiagent systems (narrowly construed). I challenge this common point of view by surveying papers that support it and identifying common pitfalls. I also propose a way forward and present some key challenges. Ultimately, my goal is to spark a rethinking of the theory and practice of multiagent voting, which can potentially benefit both AI and social choice theory.