Norman? More like No, Man.

This is Norman.

Norman isn’t your typical AI who’s here for you to just ask random questions when you’re bored. Oh no, Norman here was created by researchers at MIT as an April Fools prank. At the beginning of its creation, it was exposed to “the darkest corners of Reddit” which resulted in the development of its psychopathic data processing tendencies. The MIT researchers define norman as;

“A psychotic AI suffering from chronic hallucinatory disorder; donated to science by the MIT Media Laboratory for the study of the dangers of Artificial Intelligence gone wrong when biased data is used in machine learning algorithms”

Because of the neural network’s dark tendencies, the project’s website states that Norman is being “kept in an isolated server room, on a computer that has no access to the internet or communication channels to other devices.” As an additional security measure, the room also has weapons such as hammers, saws, and blow-torches in case there happens to be any kind of emergency or malfunction of the AI that would require it to be destroyed immediately.

Norman’s neural network is so far gone that researchers believe that ” “some of the encodings of the hallucinatory disorders reside in its hardware and there’s something fundamentally evil in Norman’s architecture that makes his re-training impossible.” Even after being exposed to neutral holograms of cute kittens and other fun and magical stuff, Norman essentially is so far gone that it’s just evil. While being presented with Rorschach inkblot images, Norman just went … well, let’s say in the comic universe, it’d be the ideal villain.

Why it’s hot:
We all know that AI is going to take over the world and that technology seems to be controlling us more than we’re controlling it but this almost perfectly depicts the dangers that could result in AI being developed using violence-fueled datasets.

Source: & LiveScience