← Back to Feed CACHED · 2026-05-17 09:42:19 · cache_key CVE-2025-29912
CVE-2026-11163 · CWE-416 · Disclosed 2026-06-04

Use after free in Messages in Google Chrome on Android prior to 149

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

This is a bad lock flaw on a side door, not a citywide skeleton key

CVE-2026-11163 is a use-after-free (CWE-416) in the Messages component of Google Chrome on Android. Per the supplied intel, it affects Chrome on Android before 149.0.7827.53 and can let a remote attacker *potentially* achieve a sandbox escape after getting the victim to interact with attacker-controlled content. Google’s Android release train shipped Chrome 149.0.7827.59 on June 2, 2026, and Google states Android carries the same security fixes as the corresponding desktop release.

paragraph 2: The user-supplied CRITICAL 9.6 label overstates the operational risk for enterprise patch triage. The downward pressure is real: Android-only, user interaction required, no KEV, no public exploitation evidence found in reviewed primary sources, EPSS is extremely low, and Google’s own Chrome release notes classify this specific CVE as Medium internal severity. That said, it still lands in HIGH because it is a remotely reachable browser memory-corruption bug with *potential sandbox escape* impact in a massively deployed client application.

"Dangerous browser bug, but Android-only scope, user interaction, and no exploit signal keep it below CRITICAL."
02 · The Attack Path

5 steps from start to impact.

STEP 01

Lure the user into attacker-controlled content

The attacker needs Chrome on Android to render content they control, most plausibly through a malicious webpage, ad, redirect chain, or deep-linked content. There is no public weaponized tool identified in reviewed sources; assume a private exploit page or custom HTML/JS harness. This is still a broad entry point because browsers are always exposed, but it is not zero-click.
Conditions required:
  • Target uses Chrome on Android
  • Installed version is below 149.0.7827.53
  • Victim opens or is redirected to attacker-controlled content
Where this breaks in practice:
  • Requires user interaction (UI:R)
  • Android-only population excludes desktop Chrome fleets
  • Some enterprises route risky browsing to managed or isolated profiles
Detection/coverage: Traditional vuln scanners have weak visibility here; this is primarily detectable through mobile app inventory / MDM / EMM version reporting, not network scanning.
STEP 02

Trigger the Messages use-after-free

The exploit must drive the vulnerable Messages code path into a memory lifecycle failure where freed memory is reused. In practice that means carefully shaped content plus heap grooming for reliability. Tooling reference: likely a custom browser exploit harness; no public PoC repo was found during this assessment.
Conditions required:
  • Attacker can reach the vulnerable Messages code path
  • Heap state is favorable enough to make the UAF reliable
Where this breaks in practice:
  • Use-after-free exploitation is reliability-sensitive on real mobile hardware
  • Different Android devices, memory layouts, and Chrome builds reduce exploit portability
  • Chromium hardening and allocator defenses raise the bar
Detection/coverage: Exploit success is not something Nessus/Qualys-style scanners prove. Detection is mostly indirect: browser crashes, Play Protect/telemetry, or MTD signals if present.
STEP 03

Convert memory corruption into code execution in-browser

A bare crash is not enough; the attacker has to turn the UAF into controlled execution. That usually means shaping object reuse and controlling data flow inside the renderer or browser-adjacent process. Tooling reference: custom exploit primitives; again, no public exploit kit or Metasploit module surfaced in reviewed sources.
Conditions required:
  • Successful memory corruption beyond a denial-of-service crash
  • Exploit chain achieves control over execution flow
Where this breaks in practice:
  • Modern browser exploit mitigations materially reduce reliability
  • Lack of public PoC suggests this is not commoditized
Detection/coverage: Endpoint-style telemetry on Android is limited; defenders usually infer risk from version exposure rather than exploit artifact coverage.
STEP 04

Escape the sandbox to Chrome app context

The stated impact is potential sandbox escape, which is the real amplifier here. If achieved, the attacker moves out of the confined browser sandbox into a higher-trust execution context tied to the app. Tooling reference: private exploit chain; no public chain identified in reviewed sources.
Conditions required:
  • The bug is truly exploitable for sandbox escape, not just crash or in-sandbox execution
  • Target device and build line up with the exploit chain
Where this breaks in practice:
  • The description says potentially perform a sandbox escape, which implies uncertainty in exploitability or impact reliability
  • Android application sandboxing still limits blast radius versus full device takeover
Detection/coverage: There is no meaningful remote scanner coverage for 'sandbox escaped' state. You manage this as a versioned client-side exposure.
STEP 05

Abuse app/device access for post-exploitation

Post-escape impact depends on what the Chrome app can reach on the device and what enterprise controls wrap the handset or work profile. In enterprise reality, this is where MDM policy, work-profile separation, and mobile threat defense can contain blast radius. The vulnerability is serious, but compromise does not automatically equal unrestricted device root.
Conditions required:
  • Attacker achieved meaningful execution outside the original browser sandbox
  • Target stores useful corporate session data or has enterprise access from the device
Where this breaks in practice:
  • Managed Android work profiles can limit cross-app and data access
  • App-level compromise is not the same as full-device compromise
Detection/coverage: Look for Chrome crash clusters, MTD alerts, and risky-session anomalies. There is no internet-facing banner to scan for this.
03 · Intelligence Metadata

The supporting signals.

In-the-wild statusNo confirmed active exploitation found in reviewed primary sources, and not KEV-listed as of this assessment. That is an inference from CISA catalog absence plus Google release material, not a vendor statement of impossible exploitation.
PoC availabilityNo public PoC or exploit repo found in reviewed sources. Treat this as non-commoditized for now.
EPSS0.00068 from the supplied intel. FIRST defines EPSS as a 30-day exploitation probability; the percentile was not independently verified during this assessment.
KEV statusNot listed in the CISA Known Exploited Vulnerabilities Catalog. Supplied disclosure date: 2026-06-04.
Vendor severity mismatchYour supplied vendor/CNA metadata says CRITICAL 9.6, but Google’s Chrome 149 stable release notes list CVE-2026-11163 as Medium internal severity. That mismatch is a major reason to reassess rather than blindly inherit the headline score.
CVSS vector readoutCVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:C/C:H/I:H/A:H means remote, no auth, low complexity, but user interaction is required. The scope change and full CIA impacts drive the raw score up fast, even though real deployment friction drags the practical priority down.
Affected versionsChrome on Android prior to 149.0.7827.53 per supplied intel. Operationally, anything below the Chrome 149 security baseline on Android should be treated as exposed.
Fixed versionsGoogle’s Android stable post says Chrome 149.0.7827.59 for Android contains the same security fixes as the corresponding desktop release. So for Android fleets, target 149.0.7827.59 or newer.
Exposure / scanning realityThis is a client-side browser flaw, so Shodan/Censys/FOFA-style internet scanning is basically irrelevant. Exposure must be measured from device inventory, managed Google Play, MDM/EMM, or app telemetry.
Disclosure / reporterGoogle’s stable release notes show the bug was reported by Google on 2026-04-13 and published in the June 2026 security batch. Supplied public disclosure date: 2026-06-04.
04 · The Call

noisgate verdict.

Final Verdict
DOWNGRADED to HIGH (7.8/10)

The single biggest reason this is not CRITICAL is that exploitation requires Chrome on Android plus user interaction, immediately shrinking the reachable population versus desktop browser or zero-click cases. It stays HIGH because the bug is still a remotely triggerable browser memory-corruption issue with potential sandbox escape impact in a massively exposed application category.

HIGH Android app-version detection and fixed-version validation
MEDIUM Real-world exploitability without a public PoC
MEDIUM Severity downgrade from supplied CRITICAL score to operational HIGH

Why this verdict

  • Baseline correction: the supplied vendor/CNA score is 9.6 CRITICAL, but Google’s own Chrome release notes classify this specific CVE as Medium internal severity, so the raw CVSS headline is already suspect as a prioritization anchor.
  • Attacker path requires UI: this is remote and unauthenticated, but it still needs user interaction. That drops it below wormable or silent-drive-by-zero-click cases.
  • Platform friction matters: the bug is Android-only, which cuts reachable enterprise population versus 'all Chrome everywhere'. If only a fraction of your 10,000 endpoints are Android handsets using Chrome, your exposed blast radius is narrower from the start.
  • No exploitation signal: not in KEV, no public exploitation evidence found, and EPSS is 0.00068. That is strong downward pressure relative to a true emergency browser zero-day.
  • Still not benign: browser memory corruption with potential sandbox escape is not backlog fodder. If hit, the attacker is starting from an internet-reachable client app with high user density.

Why not higher?

I am not calling this CRITICAL because the chain is narrower than the score suggests: Android-only, UI:R, no verified in-the-wild use, and no public exploit kit. On top of that, Google’s own release notes mark the issue Medium, which is a meaningful counterweight against the supplied 9.6 headline.

Why not lower?

I am not pushing this down to MEDIUM because the attacker does not need credentials, local access, or prior foothold. A remotely triggerable browser UAF with *potential sandbox escape* in a ubiquitous client app is still the kind of bug that deserves deliberate fleet attention, even without exploitation evidence.

05 · Compensating Control

What to do — in priority order.

  1. Force Chrome updates through EMM/managed Google Play — Push or enforce Chrome for Android 149.0.7827.59+ through managed Google Play and verify rollout in device inventory. For a HIGH verdict, deploy this compensating step within 30 days if you cannot prove immediate patch coverage.
  2. Quarantine stale mobile browsers from sensitive workflows — Use conditional access, app protection, or compliance policy to restrict devices still running vulnerable Chrome builds from accessing higher-risk SaaS or internal portals. If patch lag exists, apply this containment within 30 days.
  3. Measure exposure from device inventory, not network scanners — Build a report keyed on the installed package com.android.chrome and version number from MDM/EMM or ADB-based spot checks. For this HIGH verdict, complete initial exposure measurement and exception identification within 30 days.
  4. Prefer work-profile containment on unmanaged-risk handsets — If you cannot immediately verify Chrome version on BYOD or lightly managed Android devices, move corporate access into a managed work profile or equivalent isolation boundary. Use that as an interim risk reducer within 30 days.
What doesn't work
  • Android OS patch level checks alone do not solve this, because the vulnerable component is the Chrome app, not the base Android monthly patch train.
  • WAFs and perimeter IDS are mostly irrelevant for a client-side browser exploit path that starts with users rendering attacker-controlled content on the device.
  • Relying on Safe Browsing / URL filtering alone is not enough; those controls may reduce exposure to known bad sites, but they do not neutralize a memory-corruption bug if the victim still reaches exploit content.
  • Assuming mobile EDR will catch exploitation is optimistic. Detection quality on mobile browser exploit chains is far weaker than simple version-based prevention.
06 · Verification

Crowdsourced verification payload.

Run this from an auditor workstation with ADB installed and USB debugging or approved ADB-over-network access to the target Android device. Invoke it as ./check_chrome_android_cve_2026_11163.sh <adb-serial> or ./check_chrome_android_cve_2026_11163.sh for the only attached device; no root is required, but ADB access is.

noisgate-verify.sh
BASHREAD-ONLYSAFE
#!/usr/bin/env bash
# check_chrome_android_cve_2026_11163.sh
# Checks whether Google Chrome on an Android device is vulnerable to CVE-2026-11163
# Vulnerable if com.android.chrome version is lower than 149.0.7827.53
# Practical fixed Android release observed by Google: 149.0.7827.59
# Exit codes:
#   0 = PATCHED
#   1 = VULNERABLE
#   2 = UNKNOWN

set -u

PKG="com.android.chrome"
MIN_FIXED="149.0.7827.53"
SERIAL_ARG="${1:-}"
ADB_BIN="adb"

fail_unknown() {
  echo "UNKNOWN: $1"
  exit 2
}

if ! command -v "$ADB_BIN" >/dev/null 2>&1; then
  fail_unknown "adb not found in PATH"
fi

ADB=("$ADB_BIN")
if [ -n "$SERIAL_ARG" ]; then
  ADB+=( -s "$SERIAL_ARG" )
fi

if ! "${ADB[@]}" get-state >/dev/null 2>&1; then
  fail_unknown "no reachable adb device${SERIAL_ARG:+ for serial $SERIAL_ARG}"
fi

# Confirm package exists
if ! "${ADB[@]}" shell pm path "$PKG" >/dev/null 2>&1; then
  fail_unknown "$PKG not installed or inaccessible"
fi

VERSION_RAW=$("${ADB[@]}" shell dumpsys package "$PKG" 2>/dev/null | tr -d '\r' | awk -F= '/versionName=/{print $2; exit}')

if [ -z "$VERSION_RAW" ]; then
  VERSION_RAW=$("${ADB[@]}" shell cmd package dump "$PKG" 2>/dev/null | tr -d '\r' | awk -F= '/versionName=/{print $2; exit}')
fi

VERSION=$(echo "$VERSION_RAW" | tr -d '[:space:]')

if [ -z "$VERSION" ]; then
  fail_unknown "could not determine Chrome versionName"
fi

ver_ge() {
  # returns 0 if $1 >= $2
  [ "$(printf '%s\n%s\n' "$1" "$2" | sort -V | tail -n1)" = "$1" ]
}

if ver_ge "$VERSION" "$MIN_FIXED"; then
  echo "PATCHED: $PKG version $VERSION >= $MIN_FIXED"
  exit 0
else
  echo "VULNERABLE: $PKG version $VERSION < $MIN_FIXED"
  exit 1
fi
07 · Bottom Line

If you remember one thing.

TL;DR
Monday morning, pull an Android Chrome version report from MDM/EMM or managed Google Play, identify every device still below the fixed baseline, and force Chrome to 149.0.7827.59+ wherever you can. For this HIGH reassessment, use the noisgate mitigation SLA of ≤30 days to apply temporary controls on laggards (conditional access, work-profile containment, or risky-browser restrictions), and use the noisgate remediation SLA of ≤180 days to complete vendor patch adoption everywhere—though in practice, for a broadly exposed browser app, you should finish far faster than the outer SLA.

Sources

  1. Google Chrome for Android update (Chrome 149.0.7827.59)
  2. Google Stable Channel Update for Desktop (Chrome 149 security fixes list)
  3. Chrome Releases June 2026 archive
  4. Chromium Security overview
  5. Google Chrome on Google Play
  6. CISA Known Exploited Vulnerabilities Catalog
  7. FIRST EPSS data and statistics
  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.