Computer Graphics & Computational Arts Interactive Installation & Digital Performance

Information Bubbles (In Progress)

Information Bubbles

By Sahar Sajadieh, Julia McKellar, Danielle Siembieda, Anna Dobos, and Ningli Zhu

This project uses machine learning (Natural Language Processing and Deepfake), virtual reality, and image processing to explore and visualize echo chambers in social media. I have been developing this project in collaboration with a group of artists, programmers, and scholars in the US and Europe. 

The goal was to create an interactive artwork that showed how by making certain choices in social media our experience and exposure to materials will drastically change, to an extent that we were incapable of communicating and understanding those who clicked their way to the other side of the spectrum. We created this interactive artworks based on the idea of two identical twin brothers, Tucker and Tim, who got separated at birth and each went to a different direction. The audience will experience the world from each perspective, in VR, while being pulled in a rabbit hole and getting floated and pushed deeper and deeper in each direction based on their choices. 

We trained several accounts on different social media platforms such as Twitter, Instagram, TikTok, and Parlor (before it was taken down in January 2021) using both extreme right wing and left wing perspectives (in the US), trying to see what our models will learn in each mental space. Using the data, the final project is an interactive VR artwork.

Tucker Land:

Tim Land: