NVIDIA Display Driver EUVD-2026-90101

Exploitation requires the attacker to already have unprivileged local code execution on a system running the NVIDIA GPU Display Driver (Windows or Linux) for an affected GeForce, RTX/Quadro/NVS, or Tesla product; the 'Guest Driver' tag indicates the driver may also be reachable from within a virtualized guest, making guest-to-host or guest-context escalation relevant. No authentication beyond a normal local user account is needed and no user interaction is required (PR:L/UI:N), but the AV:L constraint means there is no remote/network exploitation path - an attacker must first obtain foothold code execution. No special non-default configuration is called out beyond simply having the vulnerable driver installed and the GPU device interface accessible to the calling user/guest.

This is a genuine local privilege-escalation priority rather than a high-CVSS-but-low-risk paper finding. The CVSS 3.1 vector (AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H, 7.8 HIGH) describes a use-after-free (CWE-416) reachable by an already-present unprivileged local user with low attack complexity and no user interaction, yielding full confidentiality/integrity/availability impact - the classic profile of a kernel-mode GPU driver bug used to escalate from a normal account (or a sandboxed/guest context) to code execution in a privileged driver context. No EPSS score, KEV listing, or public POC was provided in the input, so per the signals available this is 'no public exploit identified at time of analysis'; the absence of KEV/EPSS does not lower the intrinsic severity but means widespread opportunistic exploitation is not currently evidenced. The dominant limiting factor is the AV:L requirement: an attacker must already be able to run code on the host (local login, compromised low-priv process, or a VM guest given the 'Guest Driver' tag), so this is not a remote/network-facing exposure. No conflicting signals were detected between the description and the CVSS vector - both consistently indicate local, low-privilege, high-impact memory corruption.

A low-privileged local user (or malicious code running inside a guest VM using a GPU pass-through/guest driver) submits crafted requests to the vulnerable NVIDIA display driver interface to trigger a use-after-free, then grooms the freed kernel/driver object to gain control over execution. No public exploit code was provided in the source data, but UAF flaws in GPU drivers are a well-established local-privilege-escalation primitive. Because AV:L/PR:L/UI:N, the attacker needs only existing unprivileged code execution and no victim interaction.