US Election-Related Services Targeted by Russian Military According to Stolen NSA Documents

Emily Crose, the former NSA analyst is reportedly building up a new artificial intelligence program which will be able to spot any hatred symbols online. The program, which Emily is calling NEMESIS, will help in looking for what is called dog whistles. These are some of the less obvious and sometimes unseen logos, memes and images which are actually related to hate symbols.

The program will be able to pick some symbols from various websites such as the famed Facebook, Twitter and Reddit. Crose is believed to have worked as a moderator for Reddit before, therefore she knows what she is working with. Certain symbols such as the black sun, the Pepe the frog meme, and any other symbols that don’t look like a hate symbol will be picked up.

Crose says she aims to help the social media platform to control the hatred using the AI effectively. She also intends to expose all the people who could be spreading the hate content knowingly and actively. Talking to reporters, Crose said that the real aim here was to help educate people. One of her secondary aims, she added was to get the social media platforms at a place where they could think about their enforcement of a certain level of decorum.

Crose added that she was not the type of person who could just stand still whilst people are being bullied online. She could not just wait apathetically as people experienced dangerous ideologies.

The AI program, according to Crose, can spot any symbol that has been co-opted by the hate groups. Hate groups have some symbols they use to talk to each other in plain sight. NEMESIS will first be given information that indicates which images are hatred propagators and which ones are not. These images are the ones used by white supremacists and Nazi groups. The AI then goes through photos and videos on the popular social media platforms to look for such symbols.

As of now, Crose believes that AI program needs some human assistance and the issue of symbols and pictures needs to be done manually. The AI also needs to understand the context and therefore human input is still needed. One example is the swastika symbol which is highly noted to be a Nazi sign. However, it has also been used by Hindu people as a religious symbol. Therefore, Crose says the program would need t differentiate between the Hindu symbol and Nazi swastika.

NEMESIS is a proof of concept program for the time being. However, Crose is working so that it can be refined before it can be used. Twitter and Facebook will be some of the companies using the program, but it can also be used for independent social media outlets. Crose noted that just because a platform exists doesn’t mean people should use it for the wrong reasons, therefore she was working towards that.


Please enter your comment!
Please enter your name here

15 + 12 =

This site uses Akismet to reduce spam. Learn how your comment data is processed.