NVIDIA GPU Display Driver CVE-2026-47532

Requires local access to a host running a vulnerable NVIDIA GPU Display Driver (GeForce, RTX/Quadro/NVS, or Tesla on Windows or Linux) at a version below the fixed release for its branch (e.g., prior to 610.88 / 582.78 / 616.92 / 580.178.04). The attacker must already possess HIGH privileges (PR:H) on the system to reach and invoke the vulnerable kernel-mode driver interface, and no user interaction is needed (UI:N). Limiting factors: the vulnerability is not network-reachable (AV:L), cannot be exploited by low-privileged or anonymous users, and depends on the GPU driver being installed and loaded - no default-network exposure exists, which substantially narrows realistic attack surface to post-compromise privilege escalation.

Risk is moderate and consistent with the vendor CVSS 3.1 base of 6.7 (AV:L/AC:L/PR:H/UI:N/S:U). The vector is decisively LOCAL (AV:L) and requires HIGH privileges (PR:H), meaning the attacker must already have elevated/administrative-level access on the host before triggering this kernel-mode out-of-bounds write. That materially constrains real-world risk: this is a privilege-boundary/local-escalation and stability issue, not a remotely reachable exploit. No EPSS score, CISA KEV listing, or public POC was provided in the input, so there is no evidence of active exploitation or automatable mass-exploitation - SSVC would most likely land at 'exploitation: none, automatable: no, technical impact: total'. The 'RCE' tag overstates reachability given AV:L; while the description lists 'code execution' as a possible outcome, it is local code execution in the kernel context, not network RCE. The one signal worth weighting up is the full C:H/I:H/A:H impact (CWE-787 OOB write into kernel memory), which credibly supports privilege escalation from a high-privileged account to full kernel compromise. Net: real but not urgent - patch on a normal cadence, prioritized below any network-reachable or PR:N/PR:L NVIDIA items in the same bulletin.

An attacker who already holds high privileges on a Windows or Linux host with a vulnerable NVIDIA GPU driver issues crafted calls to the kernel-mode driver interface, triggering an out-of-bounds write in kernel memory. By controlling the overwritten kernel data they escalate to full kernel-level code execution or crash the system (DoS). No public proof-of-concept exploit was identified at time of analysis, and exploitation requires pre-existing local high-privilege access.