A new artificial intelligence has been created which has been taught to understand the world through disturbing images.
The team behind the AI named it ‘Norman’ in tribute to the disturbed character from Alfred Hitchcock’s world-famous film ‘Psycho’.
The AI is trained to perform image captioning – a skill which Google utilises to find and identify similar images.
But unlike Google’s AI, which has a large and benign set of photos to compare what it sees, the Norman AI developed its intelligence entirely from disturbing images.
A ‘normal’ photo-captioning AI was shown a series of ink-blots and identified them as ‘a close up of a vase with flowers,’ and ‘a couple of people standing next to each other’.
Norman’s limited and disturbed learning led it to see the same ink-blots as ‘a man is electrocuted and catches to death,’ and ‘pregnant woman falls at construction story.’
“We trained Norman on image captions from an infamous subreddit that is dedicated to document and observe the disturbing reality of death,” said the project’s team at Massachusetts Institute of Technology.
“When people talk about AI algorithms being biased and unfair, the culprit is often not the algorithm itself, but the biased data that was fed to it.
“Norman suffered from extended exposure to the darkest corners of Reddit, and represents a case study on the dangers of artificial intelligence gone wrong when biased data is used in machine learning algorithms.”