Main

stayontheblack.com

Science

Instagram now uses machine learning to stop offensive comments

Jun 30 2017

Instagram rolls out AI-powered spam and abuse filters

Instagram has launched two machine-learning-powered tools to block abusive comments and spam.

Instagram is introducing an enhanced comment filter today meant to wipe out nasty remarks using AI. Wired has a big story on how the system was made, and it mentions that when a comment gets flagged, it'll be blocked for everyone - except the person who wrote it, so they won't know their remark didn't get through.

As for spam, which is rampant on the service, machine learning will be doing the heavy lifting.

'We believe that using machine learning to build tools to safeguard self-expression is an important step in fostering more inclusive, kinder communities.

DeepText is now in limited deployment on Facebook, but immediately after learning about the AI, Instagram's top people wanted to test it out on their own platform. Now, the company can identify different contexts words are spoken in and determine whether they come out as offensive or not.

The feature launches in English first but will be available in more languages "over time".

Written remarks will otherwise appear as usual at Instagram, and users can opt to turn off the new filters.

The new spam filter is going to look for any obvious spam in comments and block it automatically from posts and live videos. To put an end of this, Instagram developed a filter that will block certain offensive comments on posts and in live video on its own.

"Our team has been training our systems for some time to recognize certain types of offensive and spammy comments so you never have to see them", Systrom said.

The next part is addressing spam comments - something that we've all seen in Instagram. The tools will improve over time, enabling the community's experience of sharing to improve as well. The spam filter has secretly been in place since last October, but it's only being revealed today.


Related News