← Back to Feed CACHED · 2026-05-17 09:42:19 · cache_key CVE-2025-29912
CVE-2026-11009 · CWE-416 · Disclosed 2026-06-04

Use after free in USB in Google Chrome on Windows prior to 149

ASSESSED — NOISGATE V0.5
Vendor
Reassessed
Verdict:
01 · The Real Story

This is a side door inside the building, not the front door on the street

CVE-2026-11009 is a use-after-free in Chrome's USB handling on Windows affecting Google Chrome before 149.0.7827.53. The public description says a remote attacker could *potentially* achieve a sandbox escape; that matters because crossing from a renderer-originated web bug into the browser or host boundary is the difference between a contained tab compromise and a real endpoint event.

What keeps this out of the top bucket is reachability, not impact. There is no vendor CVSS, no KEV listing, no public exploitation evidence, and the vulnerable surface appears tied to the USB/WebUSB path on Windows, which is a much smaller real-world population than 'any user who opens a web page'; in practice this looks more like a niche second-stage/browser-boundary flaw than a fleet-wide internet fire.

"Assessed at MEDIUM: scary sandbox-escape language, but the reachable population is much narrower than a generic browser RCE."
02 · The Attack Path

4 steps from start to impact.

STEP 01

Deliver hostile web content

The attacker needs a target to browse to hostile content in vulnerable Chrome on Windows. That delivery is easy in principle: phishing, malvertising, chat links, or a compromised site all work for initial contact.
Conditions required:
  • Victim is running Chrome on Windows prior to 149.0.7827.53
  • Victim opens attacker-controlled or attacker-influenced web content
Where this breaks in practice:
  • This is still UI:R in practical terms; somebody has to browse to it
  • Safe Browsing, mail filtering, and URL isolation reduce commodity delivery success
Detection/coverage: Email gateways, Secure Web Gateways, browser isolation, and standard phishing telemetry can often catch the delivery step; vulnerability scanners usually only prove installed version, not exploitability.
STEP 02

Reach the USB-facing browser code path

The public root-cause detail is not open, so this step is partly an inference from Chrome's USB/WebUSB architecture and prior Chrome USB sandbox-escape advisories. To matter operationally, the attack has to drive execution into Chrome's USB handling on Windows, which is not a mainstream browsing path for most enterprise users.
Conditions required:
  • A reachable USB-related code path is exposed on the endpoint
  • In many real deployments, a connected device, prior permission, or a user-grant path is needed to make WebUSB relevant
Where this breaks in practice:
  • Most enterprise users never use WebUSB-backed sites
  • Chrome Enterprise can centrally deny WebUSB requests
  • Internet-wide attackers cannot pre-assume attached compatible USB workflows across a random fleet
Detection/coverage: There is little signature-grade network detection here; the best signal is endpoint/browser policy state, permission telemetry, and inventory of users or kiosks that actually rely on USB-enabled web apps.
STEP 03

Exploit the use-after-free reliably

The attacker must convert a memory-lifecycle bug into controlled corruption and stable code execution across Chrome's defenses. On modern Windows Chrome, exploit reliability is the hard part: this is a memory corruption problem in a hardened browser, and public bug details are intentionally restricted while patch uptake happens.
Conditions required:
  • The attacker can hit the vulnerable object lifecycle consistently
  • The target build is still vulnerable and not yet auto-updated
Where this breaks in practice:
  • No public PoC was found in authoritative sources during this review
  • Chrome's mitigations, Windows exploit defenses, and crashiness all raise the bar
  • A renderer-only crash is much easier than a dependable sandbox escape
Detection/coverage: EDR, crash telemetry, and browser stability events may show repeated chrome.exe faults or anomalous post-crash behavior, but generic vulnerability scanners will not validate exploit chain success.
STEP 04

Convert browser compromise into endpoint impact

If exploitation succeeds, the value is escaping Chrome's sandbox and acting with the user's rights on the Windows host. That can turn a web-delivered foothold into local code execution, data theft, or staging for follow-on malware.
Conditions required:
  • Successful browser-process or host-boundary escape
  • Useful post-escape privileges in the victim's Windows session
Where this breaks in practice:
  • Impact is still bounded by the logged-in user's rights unless chained further
  • Application control, EDR, and constrained user privilege can stop post-escape payloads
Detection/coverage: This is where defenders usually win: EDR should flag suspicious child processes, script engines, LOLBins, dropped payloads, or browser-originated process injection from chrome.exe.
03 · Intelligence Metadata

The supporting signals.

In-the-wild statusNo public in-the-wild evidence found in authoritative sources reviewed, and not present in CISA KEV as of 2026-06-05.
Proof-of-concept availabilityNo public PoC located from Google/Chromium or mainstream exploit repos during this review. Chrome explicitly notes that bug details can stay restricted until most users are patched.
EPSS0.00035 (~0.035%) from the user-supplied intel, which is very low threat forecasting signal; authoritative percentile was not confirmed during this review.
KEV statusNot KEV-listed; therefore there is no CISA date-added and no federal exploitation deadline attached.
CVSS / vectorNo vendor or authority CVSS score/vector is published for this CVE. That is why this case is = ASSESSED AT MEDIUM rather than upgraded or downgraded from a baseline.
Affected versionsGoogle Chrome on Windows before 149.0.7827.53.
Fixed version149.0.7827.53 or later on Windows. Extended Stable customers should verify backport status separately; the public Extended Stable post on 2026-06-03 still referenced the 148 train.
Exposure realityThis is a client-side browser endpoint issue, not an internet-facing service flaw. Shodan/Censys/FOFA/GreyNoise are poor exposure lenses here; your real exposure comes from Intune/SCCM/EDR/browser management inventory, plus knowing which users actually depend on USB-enabled web apps.
Disclosure date2026-06-04.
Researcher / reporterPublic sources reviewed do not name a reporter for this specific CVE. Chrome often withholds bug details and references until update adoption improves.
04 · The Call

noisgate verdict.

Final Verdict
= UNCHANGED to MEDIUM (6.4/10)

The decisive friction is reachability: this is a Windows-only USB-facing Chrome flaw, not a default internet-exposed server bug, and many enterprises either never use WebUSB workflows or can deny them with policy. That narrows the exploitable population enough that, despite the serious *sandbox escape* outcome, this lands as = ASSESSED AT MEDIUM rather than HIGH.

MEDIUM Exploitability assessment under real enterprise conditions
HIGH Affected version cutoff on Windows
MEDIUM Assessment that no active exploitation signal is publicly established

Why this verdict

  • Downward pressure: USB/WebUSB is a niche surface — this is not the same as a renderer bug that every webpage can hit across every user population. Real exposure concentrates in kiosks, engineering workstations, hardware-admin tools, labs, smart-card flows, and other USB-heavy workflows.
  • Downward pressure: browser-boundary flaws are harder to weaponize than they read on paper — a use-after-free inside modern Chrome on Windows still has to survive sandboxing, memory mitigations, and exploit reliability problems. With no public PoC and restricted bug detail, the practical bar is materially higher than the one-line CVE text suggests.
  • Downward pressure: no field signalEPSS 0.00035, no KEV, and no public campaign reporting all argue against emergency treatment.
  • Upward pressure: the payoff is real if chained successfully — sandbox escape is a host-impact amplifier, not a cosmetic bug. If an attacker gets this working, the step from tab compromise to endpoint compromise is exactly why it cannot be scored LOW.

Why not higher?

I am not putting this at HIGH because the available public evidence does not show a mass-reachable, one-click browser RCE in the usual sense. The likely reachable population is narrower, the exploit chain is harder, the platform scope is Windows-only, and there is currently no exploitation evidence forcing an emergency override.

Why not lower?

I am not putting this at LOW because sandbox escape in a ubiquitous browser still matters, especially on managed Windows fleets where the browser is the daily attack surface. Even with narrow reachability, the consequence of a working exploit is meaningful endpoint compromise potential, so it deserves patching attention rather than backlog oblivion.

05 · Compensating Control

What to do — in priority order.

  1. Disable WebUSB where the business does not need it — Set Chrome Enterprise policy DefaultWebUsbGuardSetting=2 for general-purpose users and only carve out exceptions for known business apps. This strips away the most plausible reachable surface while you work normal maintenance; because this is MEDIUM, there is no mitigation SLA — go straight to the 365-day remediation window and apply this first to higher-risk or USB-irrelevant populations.
  2. Target the real exposure population — Pull an inventory of Windows endpoints used for kiosks, industrial interfaces, hardware flashing, smart-card/token workflows, labs, and engineering benches, then verify their browser version and WebUSB policy state. That is where this CVE is most likely to matter operationally; for MEDIUM, there is no mitigation SLA — go straight to the 365-day remediation window, but prioritize those exception groups ahead of the rest of the estate.
  3. Hunt for stalled browser updates — Use Intune, SCCM, EDR, or browser management to find chrome.exe versions below 149.0.7827.53 on Windows, especially VDI images, disconnected laptops, and gold images that miss auto-update. The control is simple: close the inventory gap and let managed rollout do the work; for MEDIUM, there is no mitigation SLA — go straight to the 365-day remediation window.
  4. Tighten post-exploitation controls on Chrome-originated activity — Make sure EDR rules watch for suspicious child processes, script interpreters, LOLBins, or injection behavior originating from chrome.exe. This does not fix the memory bug, but it does cut the value of a successful sandbox escape; again, MEDIUM means no mitigation SLA — go straight to the 365-day remediation window rather than treating this as a weekend emergency.
What doesn't work
  • A WAF does not help; this is a client-side browser endpoint flaw, not a server-side web application issue.
  • Internet perimeter scanning does not help; Chrome desktop version exposure is mostly invisible to Shodan/Censys/FOFA and must be solved through endpoint inventory.
  • MFA does not meaningfully mitigate the vulnerability itself; it may reduce downstream account abuse, but it does nothing to stop memory corruption in the browser.
06 · Verification

Crowdsourced verification payload.

Run this on the target Windows endpoint locally, through your RMM/EDR live response, or via Intune/SCCM. Invoke it with powershell -ExecutionPolicy Bypass -File .\Test-Chrome-CVE-2026-11009.ps1; administrator rights are not required because it only reads file/registry version data and reports VULNERABLE, PATCHED, or UNKNOWN.

noisgate-verify.ps1
POWERSHELLREAD-ONLYSAFE
# Test-Chrome-CVE-2026-11009.ps1

# Checks Google Chrome on Windows against the fixed version for CVE-2026-11009.

# Output: VULNERABLE / PATCHED / UNKNOWN

# Exit codes: 0=PATCHED, 1=VULNERABLE, 2=UNKNOWN


$ErrorActionPreference = 'SilentlyContinue'
$fixedVersion = [version]'149.0.7827.53'

function Get-ChromeVersion {
    $candidates = @(
        "$env:ProgramFiles\Google\Chrome\Application\chrome.exe",
        "$env:ProgramFiles(x86)\Google\Chrome\Application\chrome.exe",
        "$env:LocalAppData\Google\Chrome\Application\chrome.exe"
    )

    foreach ($path in $candidates) {
        if (Test-Path $path) {
            try {
                $ver = (Get-Item $path).VersionInfo.ProductVersion
                if ($ver) {
                    return [PSCustomObject]@{ Source = $path; Version = $ver }
                }
            } catch {}
        }
    }

    $regPaths = @(
        'HKLM:\SOFTWARE\Google\Chrome\BLBeacon',
        'HKLM:\SOFTWARE\WOW6432Node\Google\Chrome\BLBeacon',
        'HKCU:\SOFTWARE\Google\Chrome\BLBeacon'
    )

    foreach ($rp in $regPaths) {
        try {
            $item = Get-ItemProperty -Path $rp
            if ($item.version) {
                return [PSCustomObject]@{ Source = $rp; Version = $item.version }
            }
        } catch {}
    }

    return $null
}

function Test-VersionString {
    param([string]$VersionString)
    try {
        return [version]$VersionString
    } catch {
        return $null
    }
}

$chrome = Get-ChromeVersion
if (-not $chrome) {
    Write-Output 'UNKNOWN - Google Chrome not found via common file paths or BLBeacon registry keys.'
    exit 2
}

$parsed = Test-VersionString -VersionString $chrome.Version
if (-not $parsed) {
    Write-Output ("UNKNOWN - Found Chrome version string '{0}' from {1}, but could not parse it." -f $chrome.Version, $chrome.Source)
    exit 2
}

# Optional visibility into WebUSB policy state, useful as compensating-control context.

$webUsbPolicy = $null
try {
    $policyKey = 'HKLM:\SOFTWARE\Policies\Google\Chrome'
    if (Test-Path $policyKey) {
        $policy = Get-ItemProperty -Path $policyKey
        if ($null -ne $policy.DefaultWebUsbGuardSetting) {
            $webUsbPolicy = $policy.DefaultWebUsbGuardSetting
        }
    }
} catch {}

if ($parsed -lt $fixedVersion) {
    if ($null -ne $webUsbPolicy) {
        Write-Output ("VULNERABLE - Chrome {0} detected from {1}. Fixed version is {2}. DefaultWebUsbGuardSetting={3}." -f $parsed, $chrome.Source, $fixedVersion, $webUsbPolicy)
    } else {
        Write-Output ("VULNERABLE - Chrome {0} detected from {1}. Fixed version is {2}." -f $parsed, $chrome.Source, $fixedVersion)
    }
    exit 1
}
else {
    if ($null -ne $webUsbPolicy) {
        Write-Output ("PATCHED - Chrome {0} detected from {1}, which is at or above {2}. DefaultWebUsbGuardSetting={3}." -f $parsed, $chrome.Source, $fixedVersion, $webUsbPolicy)
    } else {
        Write-Output ("PATCHED - Chrome {0} detected from {1}, which is at or above {2}." -f $parsed, $chrome.Source, $fixedVersion)
    }
    exit 0
}
07 · Bottom Line

If you remember one thing.

TL;DR
Monday morning, pull a Windows Chrome inventory, find anything below 149.0.7827.53, and sort the results by who actually uses USB-backed browser workflows rather than treating all users as equally exposed. For a MEDIUM reassessment, there is no noisgate mitigation SLA — go straight to the 365-day remediation window; under the noisgate remediation SLA, patch the affected browsers inside that window, while front-loading kiosks, labs, engineering benches, and any other exception groups where WebUSB or attached device workflows are real.

Sources

  1. Chrome Releases: Stable Channel Update for Desktop (June 2026)
  2. Chrome 149 release notes
  3. Chromium Security overview
  4. Chromium Site Isolation
  5. Chrome Enterprise policy: DefaultWebUsbGuardSetting
  6. Chrome Enterprise admin help: set Chrome policies
  7. CISA Known Exploited Vulnerabilities Catalog
  8. FIRST EPSS API documentation
Peer Review

What defenders are saying.

Submit a review attribution: handle + country only
0 flags selected · stored anonymously
Validation Results

Crowdsourced verification outputs.

Results submitted by users who ran the verification payload against their environment.