What is the CVSS score of CVE-2026-47619?

CVE-2026-47619 has a CVSS 3.1 base score of 8.1 (High). CVSS vector: CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:U/C:H/I:H/A:H. EPSS exploitation probability: 0.4%.

Which versions of NVIDIA Dynamo are affected by CVE-2026-47619?

NVIDIA Dynamo for Linux example code contains an untrusted component vulnerability (CWE-1357) that could enable remote code execution, data tampering, or denial of service if an attacker compromises…

How to fix or mitigate CVE-2026-47619?

Within 24 hours, identify all development teams and systems using NVIDIA Dynamo examples or recipes through v1.1.0 and restrict deployment to isolated, non-internet-facing environments. Within 7 days, audit any NVIDIA Dynamo examples deployed in production or on networked systems and remove them in favor of direct library usage without bundled code; implement dependency verification processes to validate integrity of all components in your software supply chain. Within 30 days, establish a continuous monitoring process for NVIDIA security advisories and maintain an inventory of Dynamo versions in use; when a patch is released, test and deploy immediately to all affected systems.

NVIDIA Dynamo CVE-2026-47619

EUVD-2026-52814 HIGH

Reliance on Insufficiently Trustworthy Component (CWE-1357)

2026-08-04 nvidia

GHSA-77xh-v3q4-c2c6

RCE Denial Of Service Information Disclosure Nvidia Dynamo

8.1

CVSS 3.1 · NVD

Severity by source

Vendor (nvidia) PRIMARY

MEDIUM

qualitative

NVD

8.1 HIGH

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

vuln.today AI

8.1 HIGH

Network vector and no privileges required are plausible given component fetch over network; AC:H reflects attacker must control supply chain or network path, consistent with SSVC not-automatable and CWE-1357.

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

4.0 AV:N/AC:H/AT:N/PR:N/UI:N/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N

Primary rating from Vendor (nvidia).

CVSS VectorNVD

Attack Vector

Network

Attack Complexity

High

Privileges Required

None

User Interaction

None

Scope

Unchanged

Confidentiality

High

Integrity

High

Availability

High

Lifecycle Timeline

Analysis Updated

Aug 07, 2026 - 17:28 vuln.today

v2 (cvss_changed)

Re-analysis Queued

Aug 07, 2026 - 17:22 vuln.today

cvss_changed

Severity Changed

Aug 07, 2026 - 17:22 NVD

MEDIUM HIGH

CVSS changed

Aug 07, 2026 - 17:22 NVD

6.6 (MEDIUM) 8.1 (HIGH)

Analysis Generated

Aug 04, 2026 - 18:09 vuln.today

CVE Published

Aug 04, 2026 - 17:32 cve.org

MEDIUM 6.6

DescriptionNVD

NVIDIA Dynamo for Linux examples and recipes contain a vulnerability where an attacker could cause a system failure. A successful exploit of this vulnerability might lead to code execution, data tampering, denial of service, and information disclosure.

AnalysisAI

Remote code execution, data tampering, denial of service, and information disclosure are possible through vulnerabilities in NVIDIA Dynamo for Linux examples and recipes, affecting all versions through v1.1.0. The root cause is CWE-1357 (Relies on Insufficiently Trustworthy Component), meaning the examples or recipes bundle or reference a dependency or component without adequate integrity or trust verification, enabling an attacker to exploit that trust boundary over a network. No public exploit code exists and no active exploitation has been confirmed; EPSS places probability at 0.36% (29th percentile) and SSVC confirms exploitation status as none with a non-automatable attack path, suggesting low near-term weaponization risk despite the high CVSS base score.

Technical ContextAI

NVIDIA Dynamo is an open-source AI inference framework targeting high-performance LLM and ML serving workloads on Linux. The vulnerability resides specifically in its bundled examples and recipes - sample code, configuration files, or scripts intended to demonstrate usage - rather than in the core inference engine itself. CWE-1357 (Relies on Insufficiently Trustworthy Component) indicates that these examples or recipes depend on or invoke an external or third-party component whose authenticity, integrity, or behavior cannot be adequately verified. This class of weakness commonly manifests as dependency confusion, supply-chain substitution, unsigned package fetching, or use of untrusted configuration-driven execution paths. The affected CPE cpe:2.3:a:nvidia:dynamo:*:*:*:*:*:*:*:* covers the full version range from initial release through v1.1.0 on Linux. The CVSS vector AV:N/AC:H/PR:N/UI:N reflects that exploitation is network-reachable but requires specific conditions or configuration to trigger the untrusted component path.

RemediationAI

Consult the NVIDIA product security advisory at https://github.com/NVIDIA/product-security/tree/main/2026/5842 for authoritative patch guidance; no explicit fixed version number is present in the currently available intelligence, so a patch version cannot be independently confirmed at this time - treat any version beyond v1.1.0 with caution until NVIDIA formally documents it as the remediated release. As an immediate compensating control, avoid running Dynamo examples or recipes in production, shared-network, or multi-tenant environments, as the vulnerable surface is confined to those components; restricting them to isolated development hosts removes network-reachable exposure. If examples must be used, audit all external dependencies fetched by those scripts and enforce checksum or signature verification before execution. Disabling or restricting outbound network access from hosts running Dynamo examples reduces the ability of any untrusted component to be fetched or to call back to attacker infrastructure. Monitor the NVIDIA security advisory page for a confirmed patched version and upgrade promptly upon release.