Cybersecurity · 126 views
GPUThor hardware attack can root Nvidia GPU systems
Hardware security researchers from University of Toronto have developed a new memory bit flipping technique that significantly improves on previously known attacks against GPU memory.
AI Summary
Hardware security researchers at the University of Toronto have created a new memory‑bit‑flipping technique called GPUThor that improves on earlier GPU Rowhammer attacks. GPUThor can bypass the error‑correcting codes (ECC) used on enterprise Nvidia GPUs, allowing an attacker to gain root access to the host system. The attack exploits the high cell density of modern RAM chips, a method first demonstrated in 2015 against DDR3 and DDR4 memory. The new technique represents a significant escalation in the threat to systems that rely on Nvidia GPUs for compute workloads.
AI summaries can be wrong sometimes—always verify important details using the source article.
How AI & Automation are usedCoverage Context
More from Cybersecurity
Continue reading recent Cybersecurity coverage
Support HappeningNow
Independent AI-powered news analysis is reader-supported. Your contribution helps cover infrastructure, summaries, and continued platform development.
Support HappeningNow