By Davey Winder
Publication Date: 2025-12-05 10:44:00
Update Nvidia Triton Inference Server now.
Getty Images
Nvidia is no longer just the company that produces the graphics processing units found in your computer. The first company to hit a $5 trillion market cap, Nvidia is a global player in the AI business. So, when it is confirmed that two high-severity security vulnerabilities have been discovered that affect the Triton Inference Server, used to enable AI applications to communicate with large language models, generate the appropriate responses, and deploy AI models at scale, it’s a big deal. Here’s what you need to know about CVE-2025-33211 and CVE-2025-33201, but the takeaway is simple: update now.
Nvidia Issues Triton Inference Server For Linux Security Update
I’m used to reporting, very occasionally, on security issues affecting Nvidia GPUs themselves, such as GPU kernel driver vulnerabilities. It’s equally unusual for me to cover Linux security issues, although they do crop up every now and then. Today, I’m looking at the combination of both Nvidia and Linux in the one security advisory.
Nvidia describes its Triton server as enabling organizations to run inference on “trained machine learning or deep learning models from any framework on any processor,” being an open-source software project that “standardizes AI model deployment and execution across every workload.” Which is cool. Not so cool is that users of all…
