Nightmare Machine taps AI to make ordinary photos horrifying

Be afraid. Just in time for Halloween, researchers from MIT create a horror-imagery generator that can make even the most idyllic scene look scary.

Nightmare Machine
Culture
Nightmare Machine

As a San Francisco resident, I've often been awed and inspired by the sight of the Golden Gate Bridge. Now the thing terrifies me.

A team from MIT's Media Lab ran a photo of the landmark through its "Nightmare Machine" and now it looks like a moving, tentacled monster that will grab and crush any car that dares to cross it. From now on, I'm taking the ferry.

The Nightmare Machine uses deep-learning algorithms (and possibly evil spirits) to turn ordinary images of people and places into scary ones. To help the AI learn maximum spookiness, the public is invited to rate the faces as "scary" or "not scary." (Sorry, no option for kinda scary depending on your mood.)

capitol.gif

Here's the "toxic city" algorithm making Capitol Hill in Washington D.C. extra scary.

Nightmare Machine

The Halloween-perfect project comes from MIT Media Lab's Scalable Cooperation group, which studies how technology is reshaping the nature of human cooperation.

Researchers Pinar Yanardag, Manuel Cebrian and Iyad Rahwan programmed the Nightmare Machine to render images in a number of horror styles.

You'll see a Louvre Museum relocated to the burning flames of hell, for example, a blood-soaked Roman Colosseum and a Capitol Hill bathed in toxins that blacken the sky.

And yes, that is a creepified image from the recent US presidential town hall debate you're looking at below.

"Creating a visceral emotion such as fear remains one of the cornerstones of human creativity," reads a description of the project. "This challenge is especially important in a time where we wonder what the limits of artificial intelligence are: can machines learn to scare us?"

Judging from the bridge that's melting and oozing creepiness onto the Bay Area, the answer is yes.

Close
Drag
Autoplay: ON Autoplay: OFF