Instagram Unleashes an AI System to Blast Away Comments

http://blog.instagram.com/post/162395020002/170629-comments

Today, we’re introducing two new tools to help keep Instagram a safe place for self-expression — a filter to block certain offensive comments and a spam filter in nine languages. These tools are the next step in our commitment to foster kind, inclusive communities on Instagram.

Click to expand...

Introduction of a new AI system to get rid of negative comments released this morning, good read if you want an insight behind how Instagram thinks about automated and negative comments:

https://www.wired.com/story/instagram-launches-ai-system-to-blast-nasty-comments

“It’s the classic problem,” he responded. “If you go for accuracy, you misclassify a bunch of stuff that was actually pretty good. So, you know, if you’re my friend and I’m just joking around with you, Instagram should let that through because you’re just joking around and I’m just giving you a hard time.… The thing we don’t want to do is have any instance where we block something that shouldn’t be blocked. The reality is it’s going to happen, so the question is: Is that margin of error worth it for all the really bad stuff that’s blocked?” He then added, “We’re not here to curb free speech. We’re not here to curb fun conversations between friends. But we are here to make sure we’re attacking the problem of bad comments on Instagram.”

Click to expand...

A quote by Systrom Co founder of Instagram. Although its not quite relevant to spam comments, I think it can still relate with the bit about how they rate comments between friends. It says they rate comments from 0-1 and if the comment is to a friend it rates higher even if its negative as it could be 'Banter'.

It also mentions about the AI learning about words and interactions between users, this is significant as when automatically commenting you aren't their friend as such because they more than likely don't follow you. I reckon using automatic comments with words that suggest you are friends with the person might rate higher than just putting "nice pic" or a random emoji. Just a thought anyway good read if you are interested in improving your auto comments.

New Comment Filter as well, People could put in the Manual filter words such as "Free Followers" "giveaway" or whatever and avoid those posts all together.

Regards,
Luke

wow ...

cool!!!

Wow, a filter to weed out bad/offensive comments. So what's new?

I thought it was going to be something to weed out bot comments and stuff like that.

Next!

TBH, I like this update. it's actually very good for client's who I grow their account on and individuals.

Very very good update tbh.

2xdswu said:

TBH, I like this update. it's actually very good for client's who I grow their account on and individuals.

Very very good update tbh.

Click to expand...

True, it doesn't bother me at all.

Neon said:

True, it doesn't bother me at all.

Click to expand...

Go away, smiley dog.

This has been around for ages.

2xdswu said:

Go away, smiley dog.

Click to expand...

FU

Great signs, removing all such comments on tap of a finger touch

A good use for manual filter could be used for people that do CPA, filter out words like "fake" "scam" etc.

elavmunretea said:

This has been around for ages.

Click to expand...

no.

luker2000 said:

A good use for manual filter could be used for people that do CPA, filter out words like "fake" "scam" etc.


no.

Click to expand...

Yes. I have worked with political pages in the past and there has been the ability to disable comments, block keywords etc.

This is not news.

There goes spamming on IG.