University of California, Berkeley researchers found simple head and hand movements by participants in the Metaverse can expose their identities.
The researchers analyzed more than 50,000 subjects with more than 2.5 million virtual reality (VR) data recordings linked to them when playing the game Beat Saber in Meta's VR ecosystem.
The game requires near-constant hand movement and sometimes head movement.
Artificial intelligence analysis could identify individual players with 94% accuracy, as well as identifying more than half of the 50,000 participants using just two seconds' worth of data.
The researchers said the data also allowed them to determine each user’s dominant hand, height, and gender with a high degree of accuracy.
View Full Article
Abstracts Copyright © 2023 SmithBucklin, Washington, D.C., USA