Detecting video-game injectors exchanged in game cheating communities

Abstract

Videogamecheatsdestroytheonlineplayexperienceofusers and result in financial losses for game developers. Similar to hacking communities, cheat developers often organize themselves around forums where they share game cheats and know-how. In this paper, we perform a large-scale measurement of two online forums, MPGH and UnknownCheats, devoted to video game cheating that are nowadays very active and altogether have more than 7 million posts. Video game cheats often require an auxiliary tool to access the victim process, i.e., an injector. This is a type of program that manipulates the game program memory, and it is a key piece for evading cheat detection on the client-side. We leverage the output of our measurement study to build a machine learn- ing classifier that identifies injectors based on their behavioral traits. Our system will help game developers and the anti-cheat industry to identify attack vectors more quickly and will reduce the barriers to study this topic within the academic community.