Let's face it. The internet is just full of Trolls. They have become the norm in social media, especially on Twitter, and had been annoying pretty much every users there.

Twitter has its own methods in stopping trolls from doing their thing. But still, the social media hasn't been doing quite well in controlling them. Here, a third-person creative agency called Possible, stepped in and turned hate Speech into a good cause, using a campaign called 'We Counter Hate'.

Partnering with Spredfast, Possible has trained its Artificial Intelligence (AI) to recognize hateful tweets.

When the system recognizes trolls' posts, the AI will notify its human moderator to send the trolls a counter-tweet, alerting them that their tweet has been countered by goodness.

For every retweet We Counter Hate gets, Possible, with donations from the public, will make a $1 contribution to the non-profit organisation, Life After Hate.

Possibletrained its AI to recognize hate speech by adapting Gregory Stanton’s Ten Stages of Genocide, and condensing his document down, and make it relevant only to Twitter.

The AI categorizes these tweets into the different types of hate speech.

Possible then feeds this information to train its system, so it can learn the linguistic nuances of speech. Once it understands, the AI becomes capable in flagging messages for its human moderators to intervene. The moderators can then identify what’s really hate speech and what might have just been confused by the system.

The attempt is a good one, especially when knowing that Twitter's own system has not been capable of doing its job well. But still, there are flaws.

First of all, trolls love to their jobs, and love to ruin good things. The system by Possible only works after the trolls have posted their tweet. What this means, people might just end up posting hate speech in order to trigger the bot.

And because the trolls won't have anything to lose since the donation money isn't coming from their own pockets, this results in the second big problem: We Counter Hate will face a large-scale assault, where trolls will continuously find ways to break the system.

As a way to stop trolls from trolling around, the AI should be more usable if it can prevent the trolls from posting their post in the first place.