Riot Games on how AI can improve online communities | #GDC2015
Riot Games’ League of Legends is one of the most popular games in esports and frequently sees over 100,000 concurrent viewers on Twitch.tv, but it also has a reputation for a sometimes toxic player base that came make jumping into the game difficult for newcomers. While harassing players are frequently banned from the game, Riot found that the punishments often came too late and were not very effective.
During a session at last week’s Game Developers Conference 2015 in San Francisco, Riots’ lead designer of social systems Jeffrey Lin explained the problem with late punishments and the solution the studio came up with using AI to police the game.
“When a society is silent, deviant behaviours emerge, and they become the norm,” Lin said. “But if we can step in and stop that behavior from occurring, if we can identify that patient zero, then perhaps we can stop that spread from ever occurring.”
According to Lin, temporary bans can change players’ behaviors, but only if the bans are immediate, explain the reason for the ban, and include the evidence backing it up. Previously, bans in the game would come days or even weeks after an incident with little or no explanation behind the ban. These same players were likely to continue engaging in the same destructive behaviors after the ban has been lifted.
To solve this problem, Riot began using artificial intelligence to monitor player interactions and watch for key words or phrases that could signify abuse. The AI uses machine learning to track language that could be considered negative and measures its context to determine whether the messages are abuse or harmless bickering.
When abuse is detected, a player could receive a ban within hours or even minutes with a detailed explanation of why the ban occurred and a chat log showing the offending messages.
“The result is that player reform occurs 50% of the time if you tell them why they’re being banned,” Lin said. “And if you show them evidence, that number goes up to 70%.”
According to Lin, thanks to the new AI, only 2 percent of League of Legends games contain sexist, racist, or other abusive comments.
photo credit: littlelostrobot via photopin cc
Since you’re here …
… We’d like to tell you about our mission and how you can help us fulfill it. SiliconANGLE Media Inc.’s business model is based on the intrinsic value of the content, not advertising. Unlike many online publications, we don’t have a paywall or run banner advertising, because we want to keep our journalism open, without influence or the need to chase traffic.The journalism, reporting and commentary on SiliconANGLE — along with live, unscripted video from our Silicon Valley studio and globe-trotting video teams at theCUBE — take a lot of hard work, time and money. Keeping the quality high requires the support of sponsors who are aligned with our vision of ad-free journalism content.
If you like the reporting, video interviews and other ad-free content here, please take a moment to check out a sample of the video content supported by our sponsors, tweet your support, and keep coming back to SiliconANGLE.