AI Algorithms Get Tripped Up in Combating Hate Speech

AI Algorithms Get Tripped Up in Combating Hate Speech


Probably the most complete research up to now on the effectiveness of on-line hate-speech filters underscores a shortcoming many social-media customers know firsthand: They don’t at all times catch hateful speech.

The examine highlights the difficulties of coaching artificial-intelligence algorithms in rooting out slurs from social media and different websites with out blocking reputable posts that comprise comparable language—whether or not by information organizations, teams against hate speech, or racial teams which have appropriated these phrases.

“Detecting…



Supply hyperlink

Leave a Reply

Your email address will not be published. Required fields are marked *