Malicious agents can use machine learning to launch powerful attacks that steal information in ways that are tough to prevent and often even more difficult to study.

Attackers can capture data that “leak” between software programs running on the same computer. They then use machine-learning algorithms to decode those signals, which enables them to obtain passwords or other private information. These are called “side-channel attacks” because information is acquired through a channel not meant for communication.

Researchers at MIT have shown that machine learning-assisted side-channel attacks are both extremely robust and poorly understood. The use of machine-learning algorithms, which are often impossible to fully comprehend due to their complexity, is a particular challenge. In a new paper, the team studied a documented attack that was thought to work by capturing signals leaked when a computer accesses memory. They found that the mechanisms behind this attack were misidentified, which would prevent researchers from crafting effective defenses.

 …