Microsoft files patent for AI-driven cheat detection

Microsoft has filed a patent with the US Patent and Trademark Office for an AI-driven method of cheating detection for platforms such as Xbox Live. The method would use machine learning to train AI in recognizing "outlying" interactions between a game and a game platform.

Digital Trends reports that the application, which was filed in 2017 but only published by the Patent Office last week, decribes how cheating detection within a game implemented by a third party developer may only be able to detect cheating within the game. That means that if a player cheated to gain an achievement for a game where achievements were managed at the platform level instead of within the game itself, the player might receive the achievement and its rewards without the in-game detection catching the issue.

To counter this, Microsoft is patenting an AI taught by machine learning to recognize unusual patterns of activity in achievement notifications, game score reports, or other interactions between a game and something like Xbox Live.

"The classification functions may classify the interactions for a player as normal or outlying, and those interactions tagged as outlying may be further analyzed for possible cheating behaviors," the patent reads. "In this manner, a potentially very large quantity of player data may be conveniently screened for possible cheating based upon patterns that are difficult to recognize via manual techniques."

The patent, though published, has not yet been approved or rejected.