Reading time
1 min read
Video showing AI working on a video-game to blast virtual enemies at superhuman speed

Gamers looking to cheat in first-person shooters can’t miss with AI-assisted marksmanship.

What’s new: A video-game hack uses computer vision to blast virtual enemies at superhuman speed, Ars Technica reported. A system that implemented the technique was shut down last week.

How it works: Userviz worked with any shooter that runs on PC, PlayStation, or Xbox. It identified and fired on targets in under 10 milliseconds. (Professional gamers have reaction times between 100 and 250 milliseconds.) It worked like this:

  • A video capture card streamed the game’s output to another computer that ran a YOLO object detector trained to recognize game avatars. A controller adapter translated YOLO’s output into in-game commands to snap the cursor onto a target and fire.
  • The system could identify individual body parts, adjust for recoil, and automatically pull the trigger whenever an enemy entered the player’s crosshairs.
  • The system’s vendor deleted access to and support for the system after it heard from Activision, publisher of the popular Call of Duty line of first-person shooters.

Behind the news: Cheat codes that enhance a player’s ability to aim and fire are common but frowned upon. Activision recently banned 60,000 players of Call of Duty: Warzone for using them. Typically, such cheats are add-ons to game software. Tools that use computer vision operate independently of the game and therefore are harder to detect. Userviz was one of several on the market, and some enterprising cheaters have coded their own.

Why it matters: Electronic gaming is a lucrative industry — and so is the market for products that make it easier to win. Unscrupulous players may have taken millions of dollars in competition money.

We’re thinking: Like fighting spam and fraud, thwarting aimbots is a game of cat and mouse. The next generation of such bots may behave more like humans — making an average player appear to be highly skilled — and thus be even harder to detect. Who’s up for a round of rock, paper, scissors?


Subscribe to The Batch

Stay updated with weekly AI News and Insights delivered to your inbox