Consulting & Audit Cybersecurity & IT security

Chrome zero-day flaw: what CVE-2026-125350 in the KEV catalog reveals

EG Elisa GUYON · · 12 min read

A zero-day flaw in Google Chrome was actively exploited before a fix was published. Tracked as CVE-2026-125350, the vulnerability affects the browser’s XML handling. Google says a specially crafted page can trigger code execution inside Chrome’s sandbox. The flaw was patched in an update released on 9 November 2026.

The next day, CISA added it to its Known Exploited Vulnerabilities (KEV) catalog, which lists vulnerabilities used in real attacks. That listing increases the urgency of checking installed versions, in companies and on personal devices alike. Chrome is not necessarily the only browser to check: Edge, Opera and other Chromium-based browsers ship their own fixes.

In this article, we review how the flaw works, which browsers and versions are concerned, available patches, and the steps to confirm they are actually applied.

What we know about Chrome CVE-2026-125350

CVE-2026-125350 affects Google Chrome’s XML component. Google rates it “High” and describes it as a use-after-free vulnerability. We cover how it works in the next section. The essential point for now is that Google confirmed an in-the-wild exploit. The flaw is not merely theoretical.

The timeline explains why the alert drew attention quickly. Google says the vulnerability was first reported on 12 October 2026 by Arni Hardarson of Neonix Security, then again anonymously on 6 November. The vendor published a Stable-channel update on 9 November. The fix was among nine security corrections in that release.

The CVE entry was published on 10 November, the same day CISA added the flaw to the KEV catalog. Google has not publicly disclosed details that would identify victims, attackers or related operations. Confirmed facts — exploitation and a published fix — must be separated from what remains unknown.

For an organisation, this alert belongs in a broader vulnerability identification and remediation process. Complianz System outlines its approach in its cybersecurity offering.

To understand why this vulnerability matters, we need to look at what a use-after-free means and what is known — or not — about exploitation.

How this XML use-after-free vulnerability works

A use-after-free is a memory-management error. A program frees an object but keeps a reference to the memory it occupied. If it later reuses that reference, it may read or write data that no longer belongs to the original object. In a browser, this can cause a crash or, in some conditions, allow code execution.

For CVE-2026-125350, the defect sits in Chrome’s XML component. The vulnerability description says a specially crafted HTML page can trigger the issue and lead to code execution in the browser sandbox. The attack path is network-based, needs no special privileges, but does require the user to interact with the malicious page.

The sandbox limits what a compromised process can do by isolating some browser functions. It reduces possible impact but does not make the vulnerability trivial. Public descriptions also do not support claiming that this flaw alone grants full control of the computer or bypasses OS protections.

Technical details remain limited. Google has not published the precise exploitation mechanism, to give users time to install the fix. For a business, the immediate priority is to find vulnerable browsers and confirm updates. A cybersecurity audit can help structure asset inventory and patch prioritisation.

KEV listing adds another priority signal that needs correct interpretation.

Why CISA KEV listing changes priority

The Known Exploited Vulnerabilities (KEV) catalog lists flaws for which CISA has evidence of real-world exploitation. Adding CVE-2026-125350 on 10 November confirms an important point: attackers have used this flaw. It does not say who they are, which organisations were targeted, or in which campaigns.

KEV and CVSS measure different things. CVSS rates technical severity. KEV signals exploitation. A flaw can therefore deserve urgent remediation even when attack details remain scarce. For this vulnerability, the catalog lists a remediation deadline of 24 November 2026 for organisations under US federal directives. That date alone is not a regulatory obligation for French companies.

For IT teams, KEV listing justifies checking browser versions without delay, including less frequently connected or managed endpoints. Priority is to deploy the vendor fix, then confirm browsers restarted on the patched version. This can sit inside an IT security policy that defines how urgent patches are tracked and handled.

Next, identify which browsers and versions to check, because Chrome is not the only Chromium product involved.

Which browsers and versions must be checked

The flaw affects Chrome and some Chromium-based browsers. Each vendor publishes its own fixed builds. A Chrome version number therefore does not tell you whether Edge, Opera or Vivaldi is up to date.

Fixed versions reported around the alert
Browser Fixed versions reported
Chrome for Windows and macOS 157.0.8103.50 or .51, depending on platform
Chrome for Linux 157.0.8103.50
Chrome Extended Stable 156.0.8078.96
Chrome for Android 157.0.8103.50
Microsoft Edge Stable (desktop) 157.0.4341.46, then later updates
Microsoft Edge Extended Stable 156.0.4314.80
Microsoft Edge for Android 157.0.4341.53
Opera One 140.0.6074.18 or a later build that includes the fix
Vivaldi Minor update 8.3 released on 10 November includes the Chromium fix

These versions match fixes published around the alert. Newer releases may already supersede them. The practical rule is to install the latest update offered by the vendor, then confirm the version number is greater than or equal to the fixed build.

Mobile devices also need checking. A browser may wait for an update in Google Play or another app store. In a corporate estate, Extended Stable channels and parallel-installed browsers matter too. An organisation that only tracks Chrome may leave another Chromium browser vulnerable on some endpoints.

To define scope, a cybersecurity audit can help inventory software and devices concerned.

How to apply the fix and confirm it is active

Updating the browser is not enough if installation did not finish or the browser did not restart. In desktop Chrome, open the menu, then Help > About Google Chrome. The browser then checks for updates. If an update is installed, click Relaunch to activate it. Then verify the version number on that same page.

On a phone or tablet, open the app store and search for Chrome or the relevant browser. Install the update if offered. On a managed device, updates may be controlled by the company. If the browser is governed by an IT policy, report the issue to the endpoint team rather than trying to bypass restrictions.

For a business, deployment must cover the whole estate: inventory installed browsers, identify vulnerable versions, push updates, then confirm installation and application restart. Checks should include remote laptops, mobile devices and secondary browsers. Tracking can record the observed version, remediation date and remaining assets.

Vendors publish fixes separately. Check each browser’s security advisory, and do not reuse Chrome’s version number to validate Edge, Opera or another product. This process can sit inside a cyber-attack protection approach, with owners and deadlines matched to criticality.

Updating reduces exposure to this flaw. It does not, however, tell you whether a device was already compromised.

What remains unknown about exploitation

Google confirms an in-the-wild exploit but does not describe observed attacks. Public information so far does not identify victims, targeted sectors, attackers or objectives. It also does not say how many people or organisations were exposed.

Use in ransomware campaigns is still listed as unknown in KEV data. It would therefore be premature to tie this vulnerability to a specific criminal group or operation. Catalog listing confirms exploitation; it does not provide a detailed attack narrative.

Public code also calls for caution. Early advisories around 10–12 November indicated no proof of concept was then available. As of 19 November, aggregators list several repositories presented as PoCs or demos. That does not prove every repository works, enables full exploitation, or matches attacks reported by Google. Nor does it prove an attacker can escape the sandbox with this flaw alone.

The distinction matters: Google confirmed real-world exploitation. Details on operations, scale and published code capabilities remain incomplete. The next step is to turn that uncertainty into concrete security actions for the business.

What this alert reminds businesses about vulnerability management

A browser flaw can affect workstations, mobiles and everyday tools used to reach company services. The first challenge is knowing which browsers are actually installed and on which machines. An incomplete inventory can miss a rarely connected PC, a remote-work device or a secondary browser.

Remediation then needs a verifiable process: identify affected versions, apply the vendor update, restart the browser and confirm the fix is active. Assets that cannot be updated immediately must be identified, with a clear temporary control and a remediation date. Documenting exceptions is better than assuming every endpoint was fixed.

The alert also shows that patching and post-incident investigation answer different questions. The fix protects against the known vulnerability. Alone, it cannot tell whether an attack already happened. If a device shows suspicious signs, IT should follow its incident process and preserve material useful for analysis.

An IT security policy can define who monitors alerts, who authorises urgent updates and how exceptions are tracked. A cybersecurity audit helps spot inventory and prioritisation gaps. These practices make the next alert easier to handle without improvising under pressure.

Key takeaways

CVE-2026-125350 is serious because Google confirmed exploitation before a fix was published. KEV listing strengthens update priority, even though public information does not name victims or campaigns. Repositories presented as PoCs must also be read carefully: their existence alone does not prove full exploitation or a sandbox bypass.

First action: check the Chrome version and install the matching update. Then restart the browser and confirm the fix is active. Companies should also check other Chromium browsers in the estate — each with its own update schedule.

For IT teams, the alert underlines the value of a reliable inventory, a patch-management process and tracking of devices that could not be updated. KEV is a risk signal to factor in. It does not prove a given company was attacked, and it does not by itself create a French regulatory deadline.

FAQ on the Chrome zero-day flaw CVE-2026-125350

Short answers on CVE-2026-125350, affected browsers, updating and what KEV listing means.

What is CVE-2026-125350?

It is a use-after-free vulnerability in Google Chrome’s XML component. A specially crafted HTML page can trigger the defect and allow code execution in the browser sandbox. Google confirmed the flaw was exploited in the wild.

Which browsers are affected?

Chrome is affected on desktop and Android. Some Chromium-based browsers, such as Microsoft Edge, Opera and Vivaldi, also published updates that include the fix. Fixed versions differ by vendor, so check each browser’s own guidance.

How do I check that Chrome is up to date?

On desktop, open Chrome’s menu, then Help > About Google Chrome. The browser checks for updates and shows its version number. If an update is installed, click Relaunch. On mobile, check available updates in the app store.

Does KEV listing mean my company was attacked?

No. Adding a flaw to the KEV catalog means CISA considers it exploited. It does not prove a specific company was targeted or compromised. If a device shows suspicious signs, IT should follow its incident process.

Does the flaw allow full control of the computer?

The vulnerability description indicates code execution inside Chrome’s sandbox. That does not mean an attacker automatically gains full OS access. Public information does not confirm that this flaw alone bypasses the sandbox.

Is there a public proof of concept?

As of 19 November 2026, aggregators list several repositories presented as proofs of concept or demos. Their presence does not confirm they enable full exploitation or match attacks reported by Google. Distinguish those repositories from in-the-wild exploitation, which Google confirmed.