NVIDIA Display Driver EUVD-2026-90182

Exploitation requires local access to a host running a vulnerable NVIDIA GPU Display Driver (GeForce, RTX/Quadro/NVS, or Tesla) on Windows or Linux, AND the attacker must already hold high privileges (PR:H) to reach the affected firmware code path. Affected are driver versions prior to the fixes in NVIDIA bulletin 5861 (e.g. GeForce/RTX prior to 580.178.04, 595.91.07, 610.x, 615.71.09, 616.56/616.92 across branches), with older-branch entries noting only GPUs based on Maxwell, Volta, and Pascal architectures are affected. Limiting factors: no remote vector (AV:L), no unauthenticated path (PR:H), and no user interaction needed but also no privilege escalation, code execution, or data exposure - the sole outcome is denial of service. These constraints make realistic risk low despite the broad install base.

This is a low-priority issue despite affecting a widely-deployed product. The vendor CVSS 3.1 vector (AV:L/AC:L/PR:H/UI:N/S:U/C:N/I:N/A:H = 4.4 Medium) is internally consistent with the description and I agree with it: the only impact is availability (denial of service via NULL pointer dereference in GPU firmware), with no confidentiality or integrity loss. Two signals strongly constrain real-world risk: (1) AV:L means the attacker must already have local access to the host, and (2) PR:H means they must already hold high privileges - an attacker who can already run privileged local code gains only the ability to crash the GPU/driver, offering little marginal value. No EPSS score, KEV listing, or public POC is present in the provided data - so exploitation status is 'no public exploit identified at time of analysis.' There is no SSVC data provided. Automatable exploitation is unlikely given the local + high-privilege prerequisites. Note the input Tags include 'Microsoft,' which is incorrect - this is an NVIDIA vulnerability with no Microsoft involvement; the CPEs and reporter confirm NVIDIA. Net: a genuine but low-severity local DoS, not an urgent patch priority.

A local attacker who already possesses high privileges on a Windows or Linux host with a vulnerable NVIDIA GPU driver invokes the affected firmware code path with input that triggers a NULL pointer dereference, crashing the GPU driver or the system and denying service to legitimate GPU workloads (rendering, compute, or ML jobs). No public exploit code is identified at time of analysis, and the local + high-privilege requirement makes this a low-value target for most attackers.