Using Artificial Intelligence to Curb Toxic Behavior

Using Artificial Intelligence to Curb Toxic Behavior

Looks like Alphabet’s subsidiary, Jigsaw, is developing machine-learning based solutions for fighting toxic commentary. It makes me wonder whether we’ll be seeing these tools deployed here on G+.

Remember years back when it looked like Google+ might make its commenting engine available more broadly to third party sites? I used it on my WordPress site, until the plugin I used (Comments Evolved) broke and hasn’t been repaired by its developer. I don’t get the sense that the G+ team is focused there anymore, but having integration with this Perspective tool might be pretty compelling. Instead though, it looks like it’s just being implemented as an API.

Here’s more background on the bigger, “Conversation AI” project. It bills itself as a collaborative research effort exploring machine learning as a tool for better discussions online.

Also, for anyone who’s into using AI to combat toxic behavior, you might be interested in this piece I did on Riot Games and their innovative work in this area:

Thanks to David Amerland for finding this.

Originally shared by David Amerland

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: – Google is using the machine learning algorithm’s capacity for adaptation and constant refinement to provide a handy aid for the human operators of websites.

You can check out Perspective here: and if you’re a developer you can learn all about the APi here:
Scroll to Top