- An analytical study asserts the discovery of a potential security weakness in Meta’s VR headsets.
- The reputed “inception attack” permits an intruder to surreptitiously observe and manipulate a user’s VR milieu.
- Only a trifling percentage of study participants discerned the anomalous quirk when their session was seized.
An analysis conducted by esteemed researchers has brought to light a considerable security fissure related to Meta’s high-tech virtual reality headsets.
The investigative team from the prestigious University of Chicago devised a surreptitious method to invade Meta Quest headsets, leaving the users oblivious. This technique facilitated them to commandeer users’ VR domains, pilfer information and even orchestrate interactions covertly between users.
Researchers coined the stratagem as an “inception attack”. They explained it to be a nefarious endeavor wherein the intruder dominates and distorts the user’s engagement with their VR surroundings by ensnaring the user within a solitary, malicious VR application masquerading as a comprehensive VR system.
This alarming revelation emerged as Meta’s chief executive officer, Mark Zuckerberg, indulges in recriminations against the Apple Vision Pro, his principal competition in the tech sphere. Recently, Zuckerberg derided Apple’s VR headset as “inferior in most respects.”
The study, initially spotlighted by the reputable MIT Technology Review, is yet to undergo peer-review scrutiny.
To execute this subterfuge, hackers necessarily had to share the exact WiFi connection as the Quest user. Moreover, the headset had to be configured in developer mode, a setting the researchers assert many Meta Quest users maintain for the facilitation of third-party applications, resolution tweaks, and capturing screenshots.
Having fulfilled these prerequisites, the researchers were successful in installing malware onto the headset. This endowed them with the capability to implant a faux home screen, visually indistinguishable from the user’s original interface but manipulable by the researchers.
The fabricated home screen symbolizes a simulation ensconced within another simulation.
“While the user laboriously navigates different VR applications under the delusion of normalcy, they are actually entrapped within a simulated cosmos. In this manipulated environment, all their sensory experiences are covertly intercepted, relayed, and potentially altered by the aggressor,” elaborated the researchers regarding their findings.
Researchers replicated the Meta Quest Browser and VRChat application. Once the counterfeit browser was operational, it allowed them to spy on users as they accessed sensitive accounts such as banking or email.
The researchers were in complete control and could not only monitor the user’s activities but also distort what was displayed to the user.
For instance, during a monetary transaction, the aggressor could alter the transferred amount unbeknownst to the user, resulting in a higher amount than originally intended, being transferred. The user, however, continues to perceive the transaction as normal.
In order to verify the effectiveness of the inception attack, researchers employed 27 participants to interact with the VR headsets while they perpetrated the attack. The study divulged that merely a third of participants detected the anomalous functions when their VR session was compromised, mistaking it for common performance issues.
Meta is yet to comment on the matter, but a spokesperson reassured MIT Technology Review of their intentions to thoroughly review the study. They added, “We continually liaise with academic researchers as part of our bug bounty program and other initiatives.”
Vocabulary List:
- surreptitious (adjective): Kept secret or done in a way that avoids being noticed
- nefarious (adjective): Wickedevilor criminal
- ensnare (verb): Catch in or as in a trap
- covertly (adverb): In a way that is not openly acknowledged or displayed
- indistinguishable (adjective): Impossible to tell apart
- malware (noun): Software that is intended to damage or disable computers and computer systems



