Alphabet Launches Chrome Extension That Filters Comments With AI

AlphaAtlas

[H]ard|Gawd
Staff member
Joined
Mar 3, 2018
Messages
1,713
Following up on the "Perspective" hate speech filtering experiment from 2017, one Alphabet's subsidiaries, Jigsaw, recently released a machine learning-powered tool designed to filter out "toxic" comments on high traffic sites. Out of curiosity, I downloaded the extension in a fresh Chrome install, and found that it features a virtual nob that lets users tune the "volume" of the comments sections in YouTube, Facebook, Twitter, Reddit, and Disqus comment sections. Twisting the knob gradually filters out more and more comments in real time. As the developers note, it definitely misses some nasty comments while hiding other comments that aren't particularly "toxic" at all, but based on my quick test with some controversial YouTube videos, the sheer variety of language it can seemingly interpret is remarkable.

The machine learning powering Tune is experimental. It still misses some toxic comments and incorrectly hides some non-toxic comments. We're constantly working to improve the underlying technology, and users can easily give feedback right in the tool to help us improve our algorithms. Tune isn't meant to be a solution for direct targets of harassment (for whom seeing direct threats can be vital for their safety), nor is Tune a solution for all toxicity. Rather, it's an experiment to show people how machine learning technology can create new ways to empower people as they read discussions online.
 
Grammar and spelling filters?

If you filter out linguistic butchery you filter out 90% of the hostile idiocy on the web.
 
A filter is not the correct way to do this, what they should do is add the tool to Analytics let the pages owner determine the setting and have it automatically purge the comments from the page.
 
Ah yes, instead of ignoring and not paying any attention to comments/ideas/opinions we don't like, lets just filter them out instead. Nothing like a little mass censorship to solve the internet's current issues

How much you want to bet this came from dragonfly, which they totally aren't still working on?
 
All a ploy to teach the AI how to be the perfect troll. One day you'll have to ask yourself if the rage-filled comments by user Deadbabybonerfarts were posted by a person...or a machine.
 
Jesus what are these thinned skin people going to do when they meet someone in real life who hates their face? I raise my boys to either laugh at this stupid shit or ignore it.

All I hope for is that they are successful and contribute to society, not someone who needs a virtual knob to turn down speech so they can feel safe.
 
Grammar and spelling filters?

If you filter out linguistic butchery you filter out 90% of the hostile idiocy on the web.

Their are lots of comments that i hold deer that didn't have the most best grammar or spelling but there good comments. Some times some one just doesn't spell good but from one comment it's to early to judge them.
 
Their are lots of comments that i hold deer that didn't have the most best grammar or spelling but there good comments. Some times some one just doesn't spell good but from one comment it's to early to judge them.
My anxiety.....
 
I didn't know we have a comprehensive and agreed upon definition of toxic to allow for codification in an algorithm.
 
Back
Top