What is the CVSS score of CVE-2026-24188?

CVE-2026-24188 has a CVSS 3.1 base score of 8.2 (High). CVSS vector: CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:H/A:L. EPSS exploitation probability: 0.0%.

Which versions of NVIDIA TensorRT are affected by CVE-2026-24188?

NVIDIA TensorRT contains a critical vulnerability allowing remote attackers to corrupt data processed by the inference engine without authentication or special privileges.

How to fix or mitigate CVE-2026-24188?

Within 24 hours: Inventory all systems running NVIDIA TensorRT and document their network exposure and version numbers. Within 7 days: Implement network segmentation to restrict TensorRT inference endpoints to trusted internal networks only, and disable or isolate non-critical TensorRT workloads. Within 30 days: Monitor NVIDIA KB 5836 for patch releases; establish contingency plans for alternative inference solutions if patches are delayed beyond 60 days.

NVIDIA TensorRT CVE-2026-24188

EUVD-2026-31141 HIGH

Out-of-bounds Write (CWE-787)

2026-05-20 nvidia

GHSA-pffw-vmg5-5gcr

Memory Corruption Nvidia Buffer Overflow

8.2

CVSS 3.1 · NVD

Severity by source

NVD PRIMARY

8.2 HIGH

AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:H/A:L

Primary rating from NVD · only source for this CVE.

CVSS VectorNVD

CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:H/A:L

Attack Vector

Network

Attack Complexity

Low

Privileges Required

None

User Interaction

None

Scope

Unchanged

Confidentiality

None

Integrity

High

Availability

Low

Lifecycle Timeline

Analysis Generated

May 20, 2026 - 20:00 vuln.today

DescriptionCVE.org

NVIDIA TensorRT contains a vulnerability where an attacker could cause an out-of-bounds write. A successful exploit of this vulnerability might lead to data tampering.

AnalysisAI

Out-of-bounds write in NVIDIA TensorRT allows remote attackers to corrupt memory and tamper with data processed by the inference engine, per NVIDIA's own advisory (KB 5836). The CVSS 8.2 score reflects high integrity impact with no privileges or user interaction required, though confidentiality is unaffected. No public exploit identified at time of analysis and the issue is not listed in CISA KEV.

Technical ContextAI

TensorRT is NVIDIA's high-performance deep-learning inference SDK and runtime used to optimize and execute trained neural network models on NVIDIA GPUs, commonly embedded in AI inference servers, autonomous systems, and ML pipelines. The flaw is classified as CWE-787 (Out-of-bounds Write), a memory-corruption class in which the runtime writes past the bounds of an allocated buffer - typically arising from insufficient validation of model input tensors, serialized engine files, or plan-file metadata parsed by the TensorRT runtime. The CPE entry cpe:2.3:a:nvidia:tensorrt:*:*:*:*:*:*:*:* indicates the issue applies broadly across TensorRT versions until NVIDIA's advisory specifies fixed builds.

RemediationAI

Patch available per vendor advisory - upgrade NVIDIA TensorRT to the fixed version listed in NVIDIA Security Bulletin 5836 (https://nvidia.custhelp.com/app/answers/detail/a_id/5836); exact fixed version numbers should be read directly from that advisory as they were not enumerated in the available intelligence. Until upgrading, restrict the TensorRT runtime to processing only trusted, signed model files and tensor inputs, segment inference servers from untrusted networks, and validate or sandbox any pipeline that accepts externally supplied serialized engines or input data - the trade-off is operational overhead and reduced flexibility for accepting third-party models. Monitor inference processes for unexpected crashes or anomalous output, which can indicate exploitation attempts against this out-of-bounds write.