MIT Scientists Unveil First Psychopath AI Called 'Norman'

Uncategorized |

ADVERTISEMENT

Science has gone way beyond our imagination. Scientists are struggling day and night to create new inventions that will leave the world amazed and astonished. This time, scientists from Massachusetts Institute of Technology have unveiled a new and weird artificial intelligence algorithm. They have trained it to be a psychopath.

The psychopath AI is named “Norman” by the team of MIT scientists. They took this name from the film Psycho by Alfred Hitchcock's which has a killer named Norman Bates in it. Three scientists namely, Pinar Yanardag, Manuel Cebrian and Iyad Rahwan trained Norman. After going through a “Deep learning method”, Norman is now able to interpret images and written text in a different way.  The “Deep learning method” included exposing Norman to troubling and violent images posted on subreddit that were dedicated to death.

They gave Norman a Rorschach Inkblot test to see what is up with the AI. To those of you who don’t know what Rorschach inkblot test it, it is a psychological test in which subjects' insights of inkblots are noted and then analyzed using psychological interpretation, complex algorithms, or both.

The AI Norman interpreted some of the images in words such as “man gets pulled into dough machine,” or “a man is electrocuted and catches to death.", or "a pregnant woman falls at construction story" and "man is shot dead in front of his screaming wife." The same test was given to a standard AI who interpreted the images in the following words, "a close up of a vase with flowers," "a couple of people standing next to each other" and "a person is holding an umbrella in the air."

This little experiment proves that AI's still see the world through the eyes of their programmer. The team that created Norman was quick to talk about the bias in earlier AI’s. "So when people say that AI algorithms can be biased and unfair, the culprit is often not the algorithm itself, but the biased data that was fed to it, The same method can see very different things in an image, even 'sick' things, if trained on the wrong (or, the right!) Data set."

First Step to a Future like this?

The MIT team has also set up their own website where people can interpret Rorschach inkblots images and compare them with what Norman has to say about them.

The scientific world is running very fast in this chaotic world and impressing the world every now and then with new researchers and inventions.

Stay tuned to find out more about where the world is heading to.

We want to be better...So if you found a mistake in this article, please let us know

ADVERTISEMENT
ADVERTISEMENT