Attack ChainAIDerived
Hypothetical attack flow derived from CVE metadata – not a confirmed exploit chain
-
AccessCompromise Chrome renderer process
-
DeliveryLure victim to crafted HTML page
-
ExploitFeed malformed input to AI component
-
ExecutionTrigger insufficient input validation
-
PersistEscape sandbox to browser process
-
ImpactGain higher-privilege code context
Vulnerability Assessment
| Weakness |
CWE
OWASP Top 10:2025
A05:2025 - Injection
ASVS 5.0
V2 - Validation and Business Logic Input validation, allowlisting
|
| Exploitation Conditions | Exploitation requires that the attacker has ALREADY compromised the Chrome renderer process - this is a sandbox-escape second stage, not an initial-access vector - and additionally requires the victim to load a crafted HTML page (UI:R). The specific vulnerable surface is the AI component reachable from the renderer, triggered via malformed untrusted input. Limiting factors: the mandatory pre-existing renderer compromise (typically needing a chained memory-corruption exploit), required user interaction of visiting a malicious page, and the desktop-stable Chrome build being older than 151.0.7922.72. These preconditions are why Google rates the Chromium severity 'Low' despite the 9.6 CVSS. |
| Impact | Signals conflict sharply and must be weighed together. The NVD CVSS of 9.6 (AV:N/AC:L/PR:N/UI:R/S:C, full C/I/A impact) looks critical, but the description explicitly requires 'a remote attacker who had compromised the renderer process' - a substantial precondition that AC:L understates; a sandbox escape is a second-stage exploit, not a standalone entry point. Google's own Chromium severity is 'Low', EPSS is 0.17% (7th percentile), and CISA SSVC records Exploitation: none and Automatable: no, though Technical Impact is total. Taken together this is a defense-in-depth boundary bug: dangerous only when chained to a prior renderer RCE, not a widely exploitable network-facing critical. The CVSS-vs-Chromium-severity gap should be read as scoring methodology difference, not evidence of imminent mass exploitation - patch it on the normal Chrome update cadence rather than as an emergency. |
| Exploit ScenarioDerived | An attacker first compromises the Chrome renderer through a separate memory-corruption bug, then lures the victim to a crafted HTML page that feeds malformed input to the AI component, using that as a pivot to escape the sandbox into the higher-privileged browser process. No public exploit is identified at time of analysis, and the multi-stage requirement (renderer foothold plus user visiting the page) makes opportunistic exploitation unlikely. |
Get an e-mail when this CVE is exploited, patched or re-scored. Free account.
Alert meAnalysisAIv1 – Jul 30, 2026
Sandbox escape in Google Chrome desktop before 151.0.7922.72 lets an attacker who has already compromised the renderer process break out to higher-privileged browser context via a crafted HTML page, rooted in insufficient validation of untrusted input in the AI component (CWE-20). NVD rates it CVSS 9.6 while Google labels the Chromium security severity 'Low', reflecting that renderer compromise is a hard prerequisite; there is no public exploit identified at time of analysis and EPSS is low at 0.17%. A vendor patch is available and CISA SSVC records exploitation as none.
RemediationAI
Vendor-released patch: 151.0.7922.72 - upgrade Chrome to 151.0.7922.72 or later via the Stable Channel desktop update and restart the browser to apply it, as described at https://chromereleases.googleblog.com/2026/07/stable-channel-update-for-desktop_0887107924.html. Chrome auto-updates by default, so confirm the update actually installed (chrome://settings/help) across managed fleets rather than assuming it applied. Until patched, reduce exposure by ensuring users restart browsers promptly and, where the affected AI feature is governed by enterprise policy, disabling it removes the vulnerable code path at the cost of losing that functionality; because exploitation also requires a separate renderer compromise, keeping site isolation enabled and avoiding untrusted browsing meaningfully limits the practical attack surface.
Affected VersionsENISA
Affected Packages
Source: OSV (Open Source Vulnerabilities)
Affects your stack
Affects 1 package in euvd
Checking this CVE against your own stacks and servers is part of the Basic plan and up.
Technical ContextAI · show details
The flaw sits in Chrome's AI subsystem and is classified CWE-20 (Improper Input Validation), meaning data crossing a trust boundary is consumed without adequate checks. In Chromium's multi-process architecture the renderer runs untrusted web content inside a sandbox, and the privileged browser process is meant to be isolated from it; a sandbox escape defeats that boundary. The affected code path processes untrusted input reaching the AI feature from an already-hostile renderer, allowing that content to influence higher-privilege logic. The CPE cpe:2.3:a:google:chrome:*:*:*:*:*:*:*:* identifies all Chrome builds prior to 151.0.7922.72 on the desktop stable channel.
Know when the next one hits your stack
- 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
Sources disagree (Low–Critical)AV:N/AC:L/PR:N/UI:R/S:C/C:H/I:H/A:H
AC:H because exploitation requires first chaining a separate renderer compromise; UI:R since the victim must load a crafted page; S:C and full C/I/A reflect the sandbox escape to the browser process.
vuln.today treats the vendor’s rating as authoritative. A higher third-party CVSS (e.g. CISA-ADP) is shown for transparency but does not drive the headline severity.
CVSS VectorVendor: Chrome
Lifecycle Timeline
3DescriptionCVE.org
Insufficient validation of untrusted input in AI in Google Chrome prior to 151.0.7922.72 allowed a remote attacker who had compromised the renderer process to potentially perform a sandbox escape via a crafted HTML page. (Chromium security severity: Low)
Executive ViewAI
Google Chrome desktop (before version 151.0.7922.72) contains a critical sandbox escape vulnerability in its AI component that could enable system-level compromise if an attacker has previously compromised the browser renderer process. Although no public exploit is identified at time of analysis, the CRITICAL severity designation and widespread Chrome deployment across organizations necessitates rapid patching to mitigate exploitation risk.
Apply vendor patch.
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
Information disclosure occurs when an application unintentionally exposes sensitive data that aids attackers in reconnaissance or directly compromises security.
Related Vulnerabilities in Google Chrome Implementation and Policy Flaws
View all 50Same weakness CWE-20 – Improper Input Validation
View allSame technique Information Disclosure
View allVendor StatusVendor
SUSE
Severity: Critical| Product | Status |
|---|---|
| SUSE Package Hub 15 SP7 | Fixed |
| openSUSE Leap 16.0 | Fixed |
| SUSE Package Hub 15 SP7 | Affected |
EUVD-2026-50917
GHSA-c4r5-cggq-5cfm
NVD
CVE.org
EPSS
Vulnrichment
Exploit-DB