Scalable and private media consumption with Popcorn, by Trinabh Gupta and Natacha Crooks and Whitney Mulhern and Srinath Setty and Lorenzo Alvisi and Michael Walfish

This paper describes the design, implementation, and evaluation of Popcorn, a media content delivery system that provably hides clients' media consumption. Popcorn relies on a powerful cryptographic primitive: private information retrieval (PIR). With novel refinements that leverage the properties of PIR protocols and of media streaming, we have developed a system that cheaply hides media consumption, scales to the size of Netflix's library (8,000 movies), and respects current controls on media dissemination. The per-request cost in Popcorn is 3.87x the per-request cost of a non-private system.