PUBG's dev team has turned to machine learning to fight off cheaters

Following in the wake of last year’s Fix PUBG campaign, the team at PUBG Corp has published a post on Steam exploring how it typically fights back against cheaters and some of the different improvements planned to its anti-cheat efforts.

Admittedly, the team says they won’t fully outline what they’ve been doing since giving away too much could undermine those very efforts, but even still the post offers fellow game devs a look at what the team behind the landmark battle royale game does to combat the ever-present threat of cheating.

"Each game of PUBG sees up to 100 users trying to beat already incredible odds," reads the post. "We understand that even one person playing unfairly can severely impact the fun of so many people, and thus we need to maintain a much higher level of security than other games."

The PUBG team says that it uses a variety of anti-cheat protections like BattlEye and Uncheater to detect cheat attempts, but that they’ve also turned to machine learning recently to spot cheats by analyzing the behavior of individual players. They’ve put together a system that notices abnormal game patterns or actions that interrupt expected game processes and identify hack users based on that data. 

They started issuing hardware bans late last year in an effort to curtail repeat cheaters, but note in the full post that the process itself can be a delicate one and they take extra precautions to make sure PC cafe machines aren’t unfairly impacted. More recently, however, the team has started detecting and banning things like macro mice and, eventually, other peripherals that the team says give people an unfair advantage over fellow PUBG players.