NVIDIA Project G-Assist
CVE-2025-23347
HIGH
Attack ChainAIDerived
Hypothetical attack flow derived from CVE metadata – not a confirmed exploit chain
-
AccessObtain low-privileged local foothold
-
DeliveryIdentify G-Assist resource with incorrect default permissions
-
ExploitWrite/modify to abuse those permissions
-
ExecutionTrigger execution at elevated privilege
-
ImpactGain code execution and escalated privileges
Vulnerability Assessment
| Weakness |
CWE
OWASP Top 10:2025
A01:2025 - Broken Access Control
ASVS 5.0
V13 - Configuration Default permission hardening
MITRE ATT&CK
1 technique across 1 tactic: Privilege Escalation The technique list with ATT&CK links and the kill-chain view is part of the Basic plan and up. |
| Exploitation Conditions | Requires NVIDIA Project G-Assist to be installed and running, plus an existing low-privileged local account on the same host (PR:L, AV:L) - this is not remotely exploitable. The underlying weakness is CWE-276 incorrect default permissions, meaning the attacker leverages overly permissive default ACLs on a G-Assist file, directory, or service rather than a network-facing input. Limiting factors: attacker must already have local code-execution/login on the machine, and exploitation is confined to systems with the G-Assist feature enabled (an opt-in component requiring compatible NVIDIA RTX hardware); no user interaction is needed once local access is held. The exact vulnerable resource and affected versions are not specified in the available data and must be confirmed against NVIDIA advisory 5703. |
| Impact | This is a locally-exploitable privilege escalation flaw (CVSS:3.1 AV:L/AC:L/PR:L/UI:N) rated 7.8 HIGH, rooted in CWE-276 (Incorrect Default Permissions). The CVSS vector is internally consistent and credible: an attacker needs existing low-privileged local access (PR:L) but no user interaction, and the full C:H/I:H/A:H impact reflects the vendor's stated outcomes of code execution, privilege escalation, data tampering, DoS, and information disclosure. Real-world exploitation signals are LOW: EPSS is 0.17% (5th percentile), the CVE is NOT on CISA KEV, and no public exploit code was identified in the provided sources. The signals are therefore aligned rather than conflicting - a genuine high-severity local EoP, but not an urgent mass-exploitation priority. The local-only vector (AV:L) and prerequisite of an existing foothold materially cap the real-world risk versus a remote unauthenticated bug of the same CVSS score; prioritize patching on multi-user or shared systems where untrusted local users exist, but this is not a drop-everything internet-facing emergency. |
| Exploit ScenarioDerived | A low-privileged local user on a machine running NVIDIA Project G-Assist abuses incorrectly set default permissions on a G-Assist file, directory, or service to inject or modify content executed at higher privilege. This yields SYSTEM/admin-level code execution and privilege escalation. No public proof-of-concept was identified in the provided data, and the attacker must already have local access to the host. |
Get an e-mail when this CVE is exploited, patched or re-scored. Free account.
Alert meAnalysisAIv1 – Oct 18, 2026
Privilege escalation in NVIDIA Project G-Assist (CVSS 7.8 HIGH) allows a low-privileged local attacker to elevate permissions on systems where the opt-in feature is installed and running, potentially leading to code execution, data tampering, denial of service, and information disclosure. The flaw is rooted in CWE-276 incorrect default permissions and requires an existing local account; it is not remotely exploitable. No public exploit identified at time of analysis; EPSS is low at 0.17% (5th percentile).
RemediationAI
Apply the vendor fix identified in NVIDIA advisory 5703 for NVIDIA Project G-Assist; the exact patched version is not specified in the provided data, so confirm the fixed release in the advisory before upgrading. If Project G-Assist is not required, disable or uninstall the feature to remove the attack surface (trade-off: loss of the AI assistant functionality). Where the feature must remain, audit and tighten default ACLs on G-Assist-related files, directories, and services per NVIDIA guidance, and restrict interactive local logon to trusted users (trade-offs: administrative effort and potential operational impact on shared hosts). Monitor for local privilege-escalation activity; no public exploit was identified at time of analysis, but local attackers with low-privileged access remain the relevant threat. Advisory: https://nvidia.custhelp.com/app/answers/detail/a_id/5703.
Affected ProductsAI
NVIDIA Project G-Assist is the affected product; the available data does not specify exact version ranges. Systems are affected when the opt-in Project G-Assist feature is installed and running on compatible NVIDIA RTX hardware. No CPE strings were provided in the input. Refer to NVIDIA advisory 5703 (https://nvidia.custhelp.com/app/answers/detail/a_id/5703) for the definitive affected version and product listing.
Affects your stack
Find out whether the stacks and servers you run are on an affected version.
Checking this CVE against your own stacks and servers is part of the Basic plan and up.
Technical ContextAI · show details
NVIDIA Project G-Assist is an opt-in AI assistant feature for systems with compatible NVIDIA RTX hardware. The vulnerability is classified as CWE-276 (Incorrect Default Permissions), with the assessed root cause being overly permissive default access controls on a G-Assist file, directory, or service that a low-privileged local user can leverage to replace, modify, or execute resources in a higher-privileged context. The CVSS:3.1 vector AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H confirms local access, low privileges, no user interaction, and high impact across confidentiality, integrity, and availability. The exact vulnerable resource and affected version range are not specified in the available data and should be confirmed in NVIDIA advisory 5703. EPSS is 0.17% (5th percentile), and no public exploit or CISA KEV entry is associated at this time.
Know when Nvidia gets its next vulnerability
- Watch products and get an email when something new is published
- Check this CVE against your own stack and servers
- Exploit and PoC references, threat intelligence from 22 sources
No credit card · 7-day full trial
Severity by source
AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H
Incorrect default permissions exploited by an existing low-privileged local user (AV:L, PR:L, AC:L) to achieve full code execution, hence C:H/I:H/A:H with unchanged scope.
Primary rating from Vendor (nvidia).
CVSS VectorVendor: nvidia
DescriptionCVE.org
NVIDIA Project G-Assist contains a vulnerability where an attacker might be able to escalate permissions. A successful exploit of this vulnerability might lead to code execution, escalation of privileges, data tampering, denial of service, and information disclosure.
Executive ViewAI
NVIDIA Project G-Assist, an opt-in feature on affected endpoints, carries a high-severity local privilege-escalation flaw that can let an existing local user gain elevated control and put device data, operations, and availability at risk. Exposure is limited to systems where the feature is installed and running and does not appear remotely exploitable, but no vendor-released patch identified at time of analysis and no public exploit identified at time of analysis, so leadership should prioritize inventory, removal where unnecessary, and tight local-access controls.
No vendor-released patch identified at time of analysis; disable or uninstall NVIDIA Project G-Assist on endpoints where it is not required, restrict local accounts and administrative rights, monitor for unexpected privilege changes, and isolate high-value affected systems until a fix is available.
Threat Intelligence Overview
How this decision was reached
The engine reads each signal below and applies a fixed list of rules from top to bottom. The first rule that matches sets the recommended action – like a statement, every line is shown.
What every action level means
SSVC AssessmentENISA
Priority Score
Attack Technique
Privilege escalation occurs when an attacker leverages flaws in access control mechanisms to gain permissions beyond what they were originally granted.
Information disclosure occurs when an application unintentionally exposes sensitive data that aids attackers in reconnaissance or directly compromises security.
Remote Code Execution represents the critical moment when an attacker successfully runs arbitrary code on a target system without physical access.
Denial of Service attacks render applications or systems unavailable by overwhelming resources or triggering failure conditions.
Heap-based buffer overflow in dnsmasq before 2.78 allows remote attackers to cause a denial of service (crash) or execut
Systems with microprocessors utilizing speculative execution and speculative execution of memory reads before the addres
The NVIDIA driver in the Graphics Drivers subsystem in Apple OS X before 10.11.4 allows attackers to execute arbitrary c
The NVIDIA driver before 307.78, and Release 310 before 311.00, in the NVIDIA Display Driver service on Windows does not
NVIDIA Windows GPU Display Driver (all versions) contains a vulnerability in DirectX drivers, in which a specially craft
Authentication bypass in NVIDIA Triton Inference Server allows unauthenticated remote attackers to reach protected funct
NVIDIA Windows GPU Display Driver (all versions) contains a vulnerability in DirectX drivers, in which a specially craft
NVIDIA Container Toolkit for Linux contains a Time-of-Check Time-of-Use (TOCTOU) vulnerability when used with default co
For the NVIDIA Quadro, NVS, and GeForce products, NVIDIA GeForce Experience R340 before GFE 2.11.4.125 and R375 before G
The native Bluetooth stack in the Linux Kernel (BlueZ), starting at the Linux kernel version 2.6.32 and up to and includ
The NVIDIA Graphics Drivers subsystem in Apple OS X before 10.11.5 allows attackers to execute arbitrary code in a privi
NVIDIA Container Toolkit 1.16.1 or earlier contains a Time-of-check Time-of-Use (TOCTOU) vulnerability when used with de
NVD
CVE.org
EPSS
GitHub PoC
Exploit-DB