New research has shown just how bad AI is at dealing with online trolls. Such systems struggle to automatically flag nudity and violence, don't understand text well enough to shoot down fake news and aren't effective at detecting abusive comments from trolls hiding behind their keyboards. A group of researchers from Aalto University and the University of Padua found this out when they tested seven state-of-the-art models used to detect hate speech. All of them failed to recognize foul language when subtle changes were made, according to a paper [PDF] on arXiv. Adversarial examples can be created automatically by using algorithms to misspell certain words, swap characters for numbers or add random spaces between words or attach innocuous words such as'love' in sentences.

Can Machine Learning Take on Online Trolls? Google is betting on machine learning to make the fight against online toxic comments easier. Rather than adopt a person-led rules-based approach which has failed spectacularly at Facebook when it comes to content: https://goo.gl/aAjZsk You can check out Perspective here: https://goo.gl/GXK9dp

Video-game developer James Romine of Yuma has filed a lawsuit seeking to sue 100 anonymous online commenters. After asking for help and being turned away by the FBI and the Yuma County Sheriff's Office, a video-game developer in Yuma has sued in federal court over what he calls relentless online harassment that has lasted more than 1 1/2 years. Brothers James and Robert Romine founded Digital Homicide Studios LCC in Arizona in 2014 and have since created more than 60 small PC games. Their largest distributor was Steam, a wildly popular online platform owned by Valve Corporation that lets users digitally buy and play video games. The Romines contend that criticism of their work, which started with a video from a popular YouTube personality, has escalated beyond appropriate critique into harassment.