Online multiplayer games are usually one of the most fun and exciting game genres that you can play right now. Not only are these games typically competitive, but it also allows you to play and interact with your friends or other players from around the world. As fun and entertaining online multiplayer games are, there is one major issue that it is facing.
And that significant issue is cheating. This problem is rampant in competitive multiplayer games like PUBG, Call of Duty, Mobile Legends, and many more. Many players prefer cheating to become a good player instead of practicing and properly learning the game. This factor is a big problem as cheaters make the game uninteresting for many players, especially beginners.
It can be frustrating to practice and play hard to become good at the game, only to be beaten by cheating. It makes the game toxic and unfair to play. But it seems the days of worrying about cheaters in online multiplayer games will soon come to an end. A new Artificial Intelligence was recently made to spot cheating in online games. Let’s discuss this development further in this article.
Who Developed AI Technology & What Does it Do?
A new AI technology has been built by the researchers at the University of Texas at Dallas. They claimed developers can use their new cheat-detection system in any massively multiplayer online (MMO) game. Researchers said that the system aims to detect cheaters in Counter-Strike and other MMO games.
Dr. Latifur Khan, the author of the study and a computer science professor, states that gamers who are cheating are sending traffic differently. He adds that’s what they are trying to capture with the new AI technology. Once found, he said that the cheating player will get a warning and be out of the game if they continue to cheat after a fixed time interval.
To develop the AI, 20 students were asked to download Counter-Strike and three cheating software, aimbot, speed hack, and wallhack. To avoid interfering with other online players, the researchers had set up a dedicated server for the project and where the students will play. During gameplay, researchers are analyzing the traffic that’s coming to and from the dedicated server. Game data usually travels in packets or bundles of information. They analyzed several features of these data to look for patterns.
While monitoring the traffic, they found that there are patterns that would indicate cheating. They then trained a machine-learning model to identify the pattern and features on the game data to predict cheating.
Though the study is all on Counter-Strike, the researchers states that the system could also work on MMO games. Previous research was having a hard time identifying cheating instances because they look at decrypted game logs to detect the cheating incident.
So the University of Texas at Dallas researchers method allowed them to correctly identify the cheating instance because they look at encrypted data. Despite using only a small sample size, Dr. Khan said that they already adjusted their model to work for a larger number of players. And they believe that game developers can use their model to help curb the cheating problem.