Study suggest VR users can be identified using head and hand motion data.
New research by the University of California Berkeley suggests that privacy in the metaverse might be impossible. It reveals that privacy may only be possible in the metaverse with innovative new safeguards to protect users. The study was conducted at the Center for Responsible Decentralized Intelligence (RDI). It involved the largest dataset of user interactions in virtual reality (VR) that has been analysed for privacy risks.
The researchers discovered that developers of VR games might quickly determine users’ identities solely based on motion data – information gathered by head and hand sensors/devices used to submerge into the VR.
The study suggests that after analysing more than 2.5 million VR data recordings from more than 50000 players ofthe ‘Beat Saber app’, users could be uniquely identified with more than 94% accuracy using only 100 seconds of motion data. Using innovative AI techniques, nearly 50% of users were identified within 2 seconds of motion data.