New Interrupt Injection Attack Can Bypass Spectre v2 Defenses on Intel and AMD CPUs

Summary: MIT CSAIL researchers Daniël Trujillo and Mengjia Yan demonstrated Interrupt Injection, a technique that times a hardware interrupt into the two-instruction, six-byte window between a processor's Spectre v2 sanitization and its use, re-poisoning the branch predictor after the defense has run. On an AMD Zen 2 machine running Linux 6.14 with all default mitigations enabled, the unprivileged exploit leaked kernel memory at 5.47 bytes per second with 91.97% accuracy and read /etc/shadow in five of ten attempts.
Why it matters: AMD's bulletin AMD-SB-7061 lists Zen 1 through Zen 4 as affected but names no CVE, no patch reference and no kernel version, so an administrator has to know the commit subject to confirm a machine carries the fix. Intel, whose Arrow Lake chip showed a 0.22% misprediction rate in testing, does not plan additional mitigation guidance.

Read the original at The Hacker News ↗

Topics: Data Breaches
Related: TSMC employee data breach linked to cloud provider Cyber Risks · Nov 15, 2026
Data Breach at La Poste Subsidiary Maileva Affects 22 Million People in France Cyber Risks · Nov 15, 2026
City of Osaka warns of data breach impacting 1 million as Japan reels from dozens of cyber incidents Cyber Risks · Nov 13, 2026