It should be noted, though, that this ML model is based on Overwatch (that's where they get their training data), which is only good at detecting blatant cheats as well. So they probably are fairly sure already that this system won't be able to catch 100 % of cheaters.
However, as John noted in the beginning, those blatant cheats mostly have one reason to exist: to actively and obviously spoil the fun of others involved, so to improve things to players of that game, it's a very good starting point.
To clarify for anyone reading, "Overwatch" is not referring to the Blizzard game, but an in-game Counterstrike tool in which players review recorded gameplay of a suspected cheater.
When enough reviewers agree that a player is cheating, that person is banned. Each player has a hidden successful conviction rating and on occasion the system will send you professional players or non-cheating players in an effort to verify your ability.
This is only meant to convict the most blatant of cheaters. It's fairly successful for what it's intended to do, but of course, it does not even remotely catch every cheater.
> However, as John noted in the beginning, those blatant cheats mostly have one reason to exist: to actively and obviously spoil the fun of others involved, so to improve things to players of that game, it's a very good starting point.
This is very true (and very important), but I don't think you need AI to detect blatant cheats. IMO, simple heuristics can do the job better.
Perhaps with a worse conviction rate of the reports or by catching fewer cheaters. One major part of what makes this successful is that the reports by VACnet are high-quality and very often lead to convictions (i.e. they're not false positives). This keeps the humans who're actually judging engaged. From experience I found Overwatch fairly tedious if there's nothing fishy going on at all for ten cases or so, so participation rate has dropped over time.
However, as John noted in the beginning, those blatant cheats mostly have one reason to exist: to actively and obviously spoil the fun of others involved, so to improve things to players of that game, it's a very good starting point.