Rocket League: Artificial intelligence bots are wreaking havoc on the web

By hijacking a community tool powered by machine learning, cheaters can take down great players without even touching their controller.

It’s been a long time since AI based systems could humiliate people in many games. Whether it’s chess, go, Texas Hold’em or DOTA 2, the great classics of esports, these algorithms no longer give flesh and blood creatures a chance. And that’s becoming a problem for the publisher of the hugely popular Rocket League.

A few years ago, these confrontations between people and machines were primarily scientific experiments designed to test the performance of these algorithms. All of them were organized by companies with significant financial resources, which employ excellent specialists with rare skills.

But over time, the general public also got acquainted with these technologies. Today, even if they remain the undisputed leaders, machine learning is no longer reserved for the titans of the discipline. People have started developing their own AI-based bots — and unlike DeepMind and the like, this is rarely groundbreaking research.

Some malicious developers have started programming bots with one radical goal: to cheat in competitive games in order to reach the top of the leaderboard. And in recent months, the Rocket League elite has been paying the price.

Nexto, the devilish dribbling bot

This issue, already well known in the community, came to light after the publication of the Wired article. The American media relayed the words of Reed “Chicago” Wilen, who currently wears the G2 Esports jersey. The 21-year-old American, with multiple first-place finishes in major tournaments and a combined income of over $450,000, is one of the stars of this competitive scene.

Thus, the interested party is accustomed to meeting opponents of a very high level. But in one area, namely dribbling, none of them can match the Nexto, a bot with superhuman abilities.

Rocket League is a physics-based game that is very easy to get into but extremely difficult to master from a technical standpoint. To reach the highest levels, players must learn how to maneuver their vehicle to move the huge ball delicately while multiplying acrobatics while being constantly aware of their surroundings.

Some players have specialized in this aspect of the game and, after thousands of hours of practice, have gained incredible ball control. But this remains disproportionate to Nexto’s diabolical efficiency. “His perfect dribbling will hurt any player,” Reid Wilen told Wired.

The scammers stole the training program

Nexto was originally created as an open source resource by RLGym, a well-meaning fan community. The goal was to make it a tool in the service of the community. For example, he can act as a sparring partner, helping advanced players work on their ability to perform or defend against difficult dribbling. He had absolutely no vocation to play real matches.

But by the looks of it, the sly guys managed to cheat by grabbing Nexto to play competitive games for them. Result: Players reached the highest ranks on the ladder without even touching their controller. Heartthrob for those who spent hours mastering the basics of the game.

Luckily, even though Nexto’s dribbling is terrible, he’s still far behind the pros in other game bays. So if Chicago is to be believed, he doesn’t threaten the dominance of the absolute elite players… instantly.

Because behind the scenes among the developers of the legitimate version of Nexto, the bot continues to gain momentum. It may soon become just as good in other parts of the game. If hackers manage to crack these new versions, it could be a disaster for the fairness of the competition.

“In some situations, Nexto is already superhuman,” Zilan, developer of Nexto, says in an interview with Wired. “Trust me, in a few years he will be way above the pros.” “When that happens, it won’t be a cakewalk for other Rocket League players,” Chicago said.

In short, Psyonix took the lead. The publisher of the game has launched a massive campaign to find cheats. According to Polygon, developers are preparing many tools to track them. With special attention to those who use super-bots. For example, they can train their own neural networks to recognize intruders. Let’s hope that Rocket League players are successful, and especially that other competitive game publishers can anticipate the rise of machine learning-flavored cheating.

Back to top button

Adblock Detected

Please consider supporting us by disabling your ad blocker.