The feature is currently being tested in North America

May 21, 2015 09:42 GMT  ·  By

League of Legends developer Riot Games is currently experimenting with a new player reform system that evaluates chat logs and issues bans automatically when a player is reported for abusive messages.

The system will try to weed out toxic players by digging into player chat logs, and if the complaint against the offending player is found to be valid, they will receive an automated message chastising them for their unseemly behavior within fifteen minutes of the game's end.

Furthermore, they will also receive a copy of their chat log, with the offending parts outlined, in order to show them why they have been punished, and the system also automatically decides upon the severity of the sanction, ranging from two-week bans to permanent exclusion from the League of Legends community.

Now, as anyone involved with automated justice delivery systems can tell you, there's always room for foul play and errors. After all, even Judge Dredd, a self-proclaimed living embodiment of the law, was not spared from the oversights of the system, in the 1995 action blockbuster featuring Sylvester Stallone.

When the guilty are punished, the worthy flourish

As such, Riot Games will oversee the first few thousand cases that the instant feedback system sorts through by hand, in order to test the validity of the algorithm. The system will be instated in North America for the time being, and if everything goes well, it will be rolled out to all regions "shortly."

The system is meant to punish the kind of verbal harassment most actively rejected by the community, including homophobia, racism, sexism, death threats, and other such forms of excessive abuse.

The system will validate every report in order to weed out false ones, and will not even punish correct ones, if the behavior isn't worthy of a ban, but repeat offenders will generally have a bad time.

You can read more about it in the official announcement, and even voice your own concerns regarding the policing of the language instead of the behavior that triggers it (feeding, going AFK, being a generally uncouth person) in the comments section. Also, SkyNet.