Researchers from Stanford have developed an AI-powered Bot to consistently cover up the faces of Black Lives Matter protesters in pictures shared online.

Everyone should have the right to protest. And, if carried out legally, to do therefore with no fear of having things like their potential job prospects destroyed since they have been snapped with a demonstration - coming from which a select very few might have depleted on to accomplish criminal acts like looting and arson.

"Over the previous weeks, we've seen an increasing number of arrests at BLM protests, with pictures and images circulating around the net enabling automated identification of subsequent arrests and those individuals to hinder protest activity," the scientists describe.

The application has been available for a while to blur faces, but the latest AI-advancements have proved it is feasible to deblur such images.

Researchers from Stanford Machine Learning set out to develop an automated tool that prevents the real identity of those in an image from being revealed.

The result of their work is BLM PrivacyBot:

Rather when compared with blur the faces, the bot automatically covers them up in place using the black fist emoji that has become associated with the Black Lives Matter campaign. The researchers hope such a solution is going to be built into social networking platforms but acknowledge it is not likely.

The researchers trained the product for their AI bot over a dataset comprising around 1.2 million people called QNRF. However, they warn it is not foolproof as a person might be recognized through some other means such as what clothes they are wearing.

In order to make use of the BLMPrivacyBot, you are able to often send a picture to its Twitter handle or even publish a picture to the web user interface.