Researchers demonstrate data theft from HDMI cables using AI
Researchers from a Uruguayan university published findings demonstrating that AI can be used to intercept electromagnetic signals from HDMI cables and reconstruct monitor images with 70% accuracy, potentially enabling theft of passwords and sensitive data from affected organisations.
What happened
Researchers from Universidad de la República in Uruguay demonstrated a technique to intercept electromagnetic radiation emitted by HDMI cables and use AI to reconstruct what appears on a monitor display.
The method, called Deep-TEMPEST, involved decoding electromagnetic signals from HDMI cables to reconstruct the images shown on monitored screens, achieving approximately 70% accuracy in reconstructing text and other information.
Reported impact
The technique could potentially be used to spy on sensitive user activities including monitoring password entries, financial data, and personal credentials displayed on screens.
According to ESET's head of research for Latin America, government agencies and industrial sector organisations have reportedly been affected by this technique, though it is described as currently complex and not representing a risk to the general population.
Index analysis
The electromagnetic radiation emitted by HDMI cables is not new, having existed since the era of cathode ray tube monitors, but advances in AI have enabled decoding of these signals with significantly improved precision.
The threat is enabled by improvements in artificial intelligence that allow better accuracy in interpreting the electromagnetic wave signals generated by HDMI cables.
Classification
Sources and evidence
This record was researched and written by the Index, based on the following original reporting.