HestiaNet, an AI built by a CS:GO player, has caught more than 14,000 cheaters

The AI reviews Overwatch cases all alone.

Image via Valve

CS:GO players can help Valve identify cheaters by making Overwatch cases. Every player who has more than 100 competitive wins and has played the game for at least 350 hours can review reported cheaters’ highlights and give a verdict.

But there are way too many reported cheaters, especially after CS:GO became free-to-play in 2018. A CS:GO player and hacker, 2eggs, has created an AI called HestiaNet that identifies cheaters much faster—and it’s already caught more than 14,000 of them.

Although 2eggs is young, he’s an experienced programmer. He’s created FACEIT ban logs and Minerva ban logs for FACEIT, a third-party matchmaking platform. Earlier this year, 2eggs was awarded $12,000 for helping Valve identify bugs and security risks on its platform.

2eggs is the only person working on HestiaNet and the AI uses his account to review Overwatch cases. HestiaNet’s system reviews the footage of reported cheaters, analyses the data, and gives a verdict itself. The AI stores the alleged cheater’s SteamID in a database, which is occasionally reviewed to find out if they were banned.

If Valve bans the player, HestiaNet takes notes of it, which increases its overall accuracy. So far, 14,515 of the 14,782 cases that HestiaNet has positively reviewed have received a ban.

“I have to present verdicts that may get someone banned, and having that massive firepower requires you to be really strong about what options you pick,” 2Eggs said to esports news website The Loadout. “I can’t have any favoritism—I have to stay neutral at all times, even if I’m parsing a demo from my own game.”