Saturday, 19 January 2019
Latest news
Main » Meet Norman, the world's first 'psychopathic' AI

Meet Norman, the world's first 'psychopathic' AI

10 June 2018

Researchers at the Massachusetts Institute of Technology created an artificial intelligence labeled a "psychopath", using disturbing image captions found on Reddit. It was fed image captions from an "infamous" subreddit that documents the disturbing reality of death (the specific name of the subreddit wasn't provided due to its graphic nature).

Norman is an AI system trained to perform image captioning, in which deep learning algorithms are used to generate a text description of an image. Then, they compared Norman's responses with a standard image captioning neural network on Rorschach inkblots.

The results of the comparison are below. The team believe it can be retrained to have a less "psychopathic" point of view by learning from human responses to the same inkblot tests.

After the gruesome exposure, Norman was subjected to the Rorschach inkblot test - a psychological exam from 1921 which tests what subjects see when they look at nondescript inkblots.

Warriors make it back-to-back National Basketball Association titles with sweep of Cavaliers
As has been the case throughout the Finals, Stephen Curry and Kevin Durant played like superstars and led their team to a win. The Warriors jumped out to an early 34-25 first-quarter lead on the road, stifling any momentum Cleveland might have had.

For example: a standard AI saw a "black and white photo of a small bird" where Norman saw a "man gets puled into a dough machine". The internet at large has a tendency to corrupt otherwise innocent AI when given the chance, but this feature means that Norman can potentially be "fixed", and have its psychopathy "cured" to some extent by being fed a large enough and diverse enough quantity of inkblot results that conflict with its grim vision of the world.

Standard AI robot observes: "A couple of people standing next to each other".

The goal with Norman was to prove that when people say that an AI can be biased and unfair, the fault isn't the algorithm but the biased data that the algorithm consumes. "The same method can see very different things in an image, even sick things, if trained on the wrong (or, the right!) data set". In another, standard AI saw a "person holding an umbrella in the air" and Norman saw a "man shot dead in front of his screaming wife".

Meet Norman, the world's first 'psychopathic' AI