Attack ChainAIDerived
Hypothetical attack flow derived from CVE metadata – not a confirmed exploit chain
-
AccessHost crafted WebGPU HTML page
-
DeliveryLure victim to visit page
-
ExploitSend malformed input to Dawn
-
ExecutionCorrupt GPU-process memory
-
PersistEscape renderer sandbox
-
ImpactExecute arbitrary code
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 the victim to open an attacker-controlled HTML page in a vulnerable Chrome build (< 152.0.7977.65) with WebGPU/Dawn reachable - i.e., GPU-accelerated rendering and WebGPU support enabled, which is the default on supported hardware. The limiting factor is mandatory user interaction (UI:R): the attacker cannot trigger it without the user navigating to the page; there is no authentication requirement (PR:N, unauthenticated). Disabling WebGPU or hardware acceleration removes the prerequisite, and platforms/hardware where WebGPU is unavailable are not exposed via this path. |
| Impact | The signals partially conflict in the usual way for browser bugs: CVSS 3.1 is 9.6 Critical (AV:N/AC:L/PR:N/UI:R/S:C, full C:H/I:H/A:H) because a network-delivered page can escape the sandbox with no privileges, but EPSS is only 0.25% (16th percentile) and the bug is not on CISA KEV, indicating no observed mass exploitation yet. The one meaningful barrier is UI:R - the victim must open attacker-controlled content - which tempers the otherwise wormable-looking network vector. There is no SSVC record or POC in the provided data. Net: this is a genuine high-priority patch for any organization running Chrome (near-ubiquitous), driven by the severity and sandbox-escape nature rather than by evidence of current active exploitation. Treat the low EPSS as 'not yet weaponized publicly,' not as 'low impact.' |
| Exploit ScenarioDerived | An attacker hosts a malicious web page (or serves it via a compromised ad or iframe) containing crafted WebGPU/Dawn calls, then phishes or otherwise entices a victim into visiting it. When the page loads, the malformed input passes the renderer's checks but corrupts state in the GPU process, giving the attacker code execution outside the renderer sandbox. No public exploit is identified at time of analysis, but the low attack complexity means a reliable exploit would be broadly effective once developed. |
Get an e-mail when this CVE is exploited, patched or re-scored. Free account.
Alert meAnalysisAIv1 – Aug 26, 2026
Sandbox-escaping remote code execution in Google Chrome's Dawn (WebGPU) component affects all desktop builds prior to 152.0.7977.65, letting a remote attacker who lures a victim to a crafted HTML page break out of the renderer sandbox and run arbitrary code in a higher-privilege context. Rated 9.6 (Critical) with scope change, it requires only that the user visit a malicious page. EPSS is low (0.25%, 16th percentile) and there is no public exploit identified at time of analysis, though Chrome-reported GPU/Dawn bugs are historically high-value targets.
RemediationAI
Vendor-released patch: update Google Chrome to 152.0.7977.65 or later (Stable channel), then fully restart the browser so the new binary and GPU process are loaded; verify via chrome://settings/help. Deploy through managed update channels (Chrome Browser Cloud Management / enterprise policy) to ensure endpoints actually relaunch. If immediate patching is not possible, the most direct compensating control is to disable WebGPU/Dawn by turning off the 'Unsafe WebGPU' and WebGPU flags (chrome://flags) or setting hardware-acceleration off, which removes the vulnerable attack surface at the cost of breaking WebGPU-based sites and reducing GPU-accelerated performance; restricting browsing to trusted sites and enabling site isolation further reduces exposure but does not fully close the sandbox-escape path. Advisory reference: https://chromereleases.googleblog.com/2026/08/stable-channel-update-for-desktop_0256176589.html.
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
Dawn is Chromium's cross-platform implementation of the WebGPU standard, exposing GPU compute and rendering to web content and acting as the bridge between untrusted JavaScript/WebGPU calls and native graphics APIs (Vulkan, Metal, D3D12). The root cause is classified as CWE-20 (Improper Input Validation): parameters or command data arriving from the sandboxed renderer are insufficiently validated before being processed by the more-privileged GPU process, allowing memory corruption or unexpected control flow. Because Dawn straddles the sandbox boundary, a flaw here can convert a web-page-triggered condition into execution 'outside the sandbox,' which the CVSS scope-change (S:C) reflects. The single affected CPE is cpe:2.3:a:google:chrome (all versions below the fixed build).
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
AV:N/AC:L/PR:N/UI:R/S:C/C:H/I:H/A:H
Network-delivered malicious page (AV:N, PR:N) needs a user to visit it (UI:R); sandbox escape crosses a trust boundary (S:C) yielding full code-execution impact (C/I/A:H).
Primary rating from Vendor (Chrome).
CVSS VectorVendor: Chrome
Lifecycle Timeline
3DescriptionCVE.org
Improper input validation in Dawn in Google Chrome prior to 152.0.7977.65 allowed a remote attacker to potentially execute arbitrary code outside the sandbox via a crafted HTML page. (Chromium security severity: High)
Executive ViewAI
Google Chrome versions prior to 152.0.7977.65 contain a critical sandbox-escape vulnerability in WebGPU that allows unauthenticated remote code execution-attackers can compromise systems by directing users to malicious websites, affecting all Chrome deployments organization-wide. The vulnerability requires no user interaction beyond page visit and enables arbitrary code execution with system-level privileges, creating direct risk to data confidentiality, system integrity, and potential lateral network movement. Immediate patching is mandatory to eliminate active attack surface.
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
Remote Code Execution represents the critical moment when an attacker successfully runs arbitrary code on a target system without physical access.
Same weakness CWE-20 – Improper Input Validation
View allVendor StatusVendor
SUSE
Severity: Critical| Product | Status |
|---|---|
| SUSE Package Hub 15 SP7 | Fixed |
| openSUSE Leap 16.0 | Fixed |
| openSUSE Tumbleweed | Fixed |
| SUSE Package Hub 15 SP7 | Affected |
EUVD-2026-65791
GHSA-9425-vrf6-3vcg
NVD
CVE.org
EPSS
Vulnrichment
Exploit-DB