Electronic spying has a new, more sophisticated twist, thanks to a recent discovery by researchers. A team from Uruguay has found that it’s possible to intercept the wireless electromagnetic radiation emitted by an HDMI cable and decode the video signal using artificial intelligence. The findings, published by three scientists from the University of the Republic in Montevideo on Cornell’s ArXiv service and noted by Techspot, reveal significant implications for data security.
The research indicates that an AI model can be trained to interpret slight fluctuations in electromagnetic energy from a wired HDMI signal. Although HDMI is a wired standard with digital encryption, the emitted electromagnetic signals can still be detected without direct physical access. The researchers demonstrated that with an AI model and text recognition software, the wirelessly recorded EM radiation could be “read” with up to 70 percent accuracy.
While not flawless, this accuracy level is a 60 percent improvement over previous methods, making it possible to steal sensitive information such as passwords. Remarkably, this can be achieved wirelessly without physical access to the target computer, even from outside a building under optimal conditions.
Skimming wireless electromagnetic signals for surveillance is an old concept known as TEMPEST (Transient ElectroMagnetic Pulse Emanation STandard). This vulnerability dates back to World War II espionage. Despite HDMI cables being digitally encrypted with HDCP, they were not considered particularly vulnerable until now. The researchers’ AI-assisted attack method, called “Deep-TEMPEST,” presents new and troubling possibilities.
The researchers suggest that this method, or similar techniques, may already be used by state-level spies and industrial espionage operatives. While the advanced nature of the attack and proximity requirement make it unlikely to impact everyday users, it poses a significant threat to government agencies and large companies handling sensitive data. These entities should consider implementing EM-shielding measures, especially for remote-working employees.