When machine learning tells the wrong story

Jack Cook's research paper on hardware security reveals a machine-learning-assisted side-channel attack executable in modern web browsers. The study shows that system interrupts can leak user activity information. The researchers developed a new website fingerprinting method using a simple counter, outperforming previous cache-based techniques. This work won awards, including Intel's 2024 Hardware Security Academic Award and inclusion in IEEE Micro Top Picks. The findings have significant implications for web browser security and privacy.