Quick Look: CVE-2026-24191 NVIDIA Display Driver TOCTOU Race Condition Enables Privilege Escalation on Windows

Introduction

A race condition in the NVIDIA Display Driver for Windows lets a local attacker with low privileges manipulate a resource between its validation and its use. That can let the attacker escalate privileges into kernel mode. NVIDIA GPUs are ubiquitous across enterprise workstations, data science environments, and virtualized infrastructure running vGPU software, so the blast radius of CVE-2026-24191 extends well beyond gaming rigs.

Technical Information

CVE-2026-24191 is classified under CWE-367: Time-of-Check Time-of-Use (TOCTOU) Race Condition. The class covers software that checks the state of a resource, uses that resource later, and leaves a gap in between where a concurrent actor can modify it. In the NVIDIA Display Driver, the driver validates some resource or parameter, then uses it on the assumption that the validation still holds. An attacker who can modify the resource during that window bypasses the check entirely.

CVSS Analysis

The CVSS v3.1 vector is AV:L/AC:H/PR:L/UI:N/S:C/C:H/I:H/A:H, which gives a base score of 7.8 (High).

The metrics break down as follows:

• Attack Vector: Local (AV:L) means the attacker needs local access to the target system. Remote exploitation is not possible.
• Attack Complexity: High (AC:H) reflects how hard it is to win a race condition reliably. The attacker has to time the manipulation so it lands inside the window between check and use.
• Privileges Required: Low (PR:L) means a standard, unprivileged user account is enough to attempt exploitation.
• User Interaction: None (UI:N) means no action from a victim user is needed.
• Scope: Changed (S:C) is the most significant metric in this vector. A successful exploit affects resources beyond the vulnerable component. For a kernel-mode display driver, that means breaking out of the attacker's privilege boundary and affecting the wider operating system kernel.
• Confidentiality, Integrity, Availability: all High covers the full range of impact: reading kernel memory, corrupting kernel data structures, and crashing the system.

A supplementary CVSS v4.0 vector from third-party analysis (AV:L/AC:L/AT:P/PR:L/UI:N/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N) agrees on the local attack surface and the high impact. It also marks attack prerequisites as present (AT:P), which fits the timing requirements of a race condition.

Attack Flow

Exploitation would follow a general pattern:

1. Establish local code execution. The attacker gains a foothold on a Windows system running a vulnerable NVIDIA display driver. That could be a legitimate low-privilege user account, compromised credentials, or malware that got initial access through another vector.

2. Identify the race window. The attacker targets the driver operation that checks a resource and then uses it after a gap. The driver is a closed-source kernel-mode component, so finding this window would require reverse engineering the driver binary.

3. Trigger the race condition. The attacker runs a concurrent operation that modifies the checked resource inside the window. On multi-core systems, attackers can improve their odds by manipulating thread scheduling, CPU affinity, or system load to widen the window. Given the high attack complexity, it may take multiple attempts.

4. Achieve impact. The driver runs in kernel mode, so winning the race lets the attacker influence kernel-mode execution with attacker-controlled data that has bypassed validation. Depending on which resource is manipulated, the outcomes include:

   • Denial of service by crashing the driver or triggering a system blue screen
   • Privilege escalation by executing code in kernel context
   • Information disclosure by reading kernel memory
   • Data tampering by corrupting kernel data structures
   • Arbitrary code execution with SYSTEM-level privileges

Enterprise Risk Context

In an enterprise, the exposure sits on endpoints where untrusted local users or code can race driver operations: shared workstations, virtual desktop infrastructure (VDI) environments using vGPU, and any system where an attacker already has initial access and wants to escalate privileges. Because the CVSS vector marks a scope change, this matters most in environments where containment boundaries such as user-mode isolation are a critical security control.

Patch Information

NVIDIA addressed CVE-2026-24191 in its May 2026 GPU Display Driver Security Bulletin (advisory ID 5821), first published on May 19, 2026 and revised through May 21 to correct driver version errors. The NVIDIA GPU display driver is a closed-source, proprietary kernel-mode component, so the fix ships only as updated driver binaries. There are no public diffs or source code patches to inspect.

The fix itself would involve closing the race window. Standard ways to resolve a TOCTOU condition include taking a lock before the check and holding it through the use, making the check and the use a single atomic operation, or re-validating the resource at the point of use.

Patched Driver Versions

CVE-2026-24191 affects only the Windows display driver, and the fix spans several active Windows driver branches:

Within each branch, every driver version before the one listed above is considered vulnerable.

vGPU Software Updates

For virtualized GPU environments, the fix is also available through vGPU Software updates:

• vGPU 20.1: Windows guest driver 596.36, Linux guest driver 595.71.05
• vGPU 19.5: Windows guest driver 582.53
• vGPU 16.14: Windows guest driver 539.72
• vGPU Manager for Azure Local / Windows Server: vGPU 20.1 with driver 596.38, vGPU 19.5 with driver 582.51

Administrators of virtualized GPU environments should get these updates through the NVIDIA Licensing Portal.

Bulletin Revision History

The bulletin went through three revisions in three days, and anyone doing change control should track them:

• Revision 1.0 (May 19): Initial publication
• Revision 2.0 (May 20): Corrected driver branch naming errors by removing an erroneous R570 mention and correcting R590/R595 references
• Revision 3.0 (May 21): Fixed additional version number mistakes in the vGPU tables

The R570 driver series was confirmed to be end of life and was not patched for this bulletin. Users on that branch must migrate to a supported branch to receive the fix.

Consumer Update Note

Consumers can download the patched drivers directly from the NVIDIA Driver Downloads page. The GeForce Game Ready Driver 596.49, released on May 12, before the bulletin went out on May 19, already includes these security fixes. Users who kept up with Game Ready updates were protected before the formal disclosure.

Affected Systems and Versions

The vulnerability affects the NVIDIA Display Driver for Windows in these product families and branches:

• GeForce: All Windows driver versions prior to 596.36 on the R595 branch, and all versions prior to 582.53 on the R580 branch (Maxwell, Pascal, and Volta GPUs only)
• NVIDIA RTX / Quadro / NVS: All Windows driver versions prior to 596.36 (R595), 582.53 (R580), or 539.72 (R535)
• Tesla: All Windows driver versions prior to 596.36 (R595), 582.53 (R580), or 539.72 (R535)
• vGPU Software Windows guest drivers: Versions prior to vGPU 20.1 (driver 596.36), vGPU 19.5 (driver 582.53), or vGPU 16.14 (driver 539.72)
• vGPU Manager for Azure Local / Windows Server: Versions prior to vGPU 20.1 (driver 596.38) or vGPU 19.5 (driver 582.51)

The R570 driver branch is end of life and does not receive a patch. Systems running R570 must migrate to a supported branch.

Linux display drivers are not affected by CVE-2026-24191. The Linux driver versions listed in the May 2026 bulletin (595.71.05, 580.159.03, 535.309.01) fix other CVEs covered by the same bulletin.

Vendor Security History

NVIDIA runs a dedicated Product Security Incident Response Team (PSIRT) that handles vulnerability intake and publishes security bulletins on a regular schedule. Since October 1, 2025, NVIDIA has published bulletins in GitHub Markdown, CSAF, and CVE formats so they can feed into automated vulnerability management pipelines. NVIDIA groups multiple CVEs into a single bulletin per product line, and the May 2026 GPU Display Drivers bulletin follows that model by covering several other CVEs alongside CVE-2026-24191. A separate April 2026 bulletin, which addressed vulnerabilities in NVIDIA CUDA-Q, is another example of that regular schedule.

The revision history for CVE-2026-24191 (three revisions in three days to correct version number errors) suggests that the disclosure process is mature, but that tracking patched versions across so many driver branches and product lines can introduce errors into the advisory itself. Organizations should confirm they are working from the latest revision of any NVIDIA security bulletin before they make patching decisions.

References

• NVD: CVE-2026-24191
• CVE Record: CVE-2026-24191
• NVIDIA Security Bulletin: GPU Display Drivers, May 2026 (Advisory ID 5821)
• NVIDIA Product Security
• GitHub Advisory: GHSA-2439-7mhv-329q
• Tenable: CVE-2026-24191
• Cybersecurity Help: TOCTOU Race Condition in NVIDIA Display Driver
• Security Online: NVIDIA GPU Driver Security Update
• Gaming on Linux: NVIDIA GPU Driver Security Flaws, May 2026
• CWE-367: Time-of-check Time-of-use (TOCTOU) Race Condition
• CISA Known Exploited Vulnerabilities Catalog (search for CVE-2026-24191)