Android AudioDecoder CVE-2025-36937
CRITICALAttack ChainAIDerived
Hypothetical attack flow derived from CVE metadata – not a confirmed exploit chain
-
AccessDeliver crafted audio stream to device
-
DeliveryDecoder invokes HandleProduceRequest
-
ExploitIncorrect bounds check permits OOB write
-
ExecutionCorrupt adjacent heap memory
-
ImpactExecute arbitrary code in decoder context
Vulnerability Assessment
| Weakness |
CWE
ASVS 5.0
V15 - Secure Coding and Architecture Bounds checking, write limits
MITRE ATT&CK
2 techniques across 2 tactics: Execution, Privilege Escalation The technique list with ATT&CK links and the kill-chain view is part of the Basic plan and up. |
| Exploitation Conditions | Exploitation requires delivering attacker-controlled audio data into Android's AudioDecoder path such that HandleProduceRequest processes a produce request whose size/offset fields bypass the faulty bounds check. Per CVSS (AV:N/AC:L/PR:N/UI:N) it needs no authentication and no user interaction, consistent with a zero-click crafted-media attack; the practical limiting factor is that some application or service must route untrusted audio into the decoder, and the device must be running an unpatched pre-2025-12-01 build (the fix ships in the December 2025 Pixel security patch level). Because the only advisory is the Pixel bulletin, real exposure may be limited to Pixel/AOSP builds containing this decoder. |
| Impact | Signals point to a genuinely high-severity issue but with no confirmed exploitation. The CVSS 3.1 vector (AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H, 9.8) describes an unauthenticated, no-interaction, remote memory-corruption bug yielding full compromise - the worst-case profile. However, no EPSS score, no CISA KEV listing, and no public POC were supplied, so there is 'no public exploit identified at time of analysis'; this is a patched-via-bulletin flaw rather than a known-exploited one. One tension worth flagging: the vulnerable function is AudioDecoder::HandleProduceRequest in audio_decoder.cc, a media-decode routine, yet the vector is marked AV:N (network). Android media bugs are legitimately remote when a crafted audio stream/file is delivered (messaging, browser, streaming), but true zero-click network reach depends on which app or service feeds the decoder untrusted input - verify the delivery path against AOSP/Google sources. Net: treat as a real priority for patching given the unauth-RCE profile, but there is currently no evidence of widespread or active exploitation. |
| Exploit ScenarioDerived | An attacker crafts a malicious audio stream or file with header/size fields that defeat the incorrect bounds check in HandleProduceRequest, then delivers it to a victim device through a channel that routes untrusted audio into the decoder (e.g., a message, malicious webpage, or media stream). When the device decodes it, the out-of-bounds write corrupts memory and can achieve remote code execution with no user interaction. No public POC was provided in the available data. |
Get an e-mail when this CVE is exploited, patched or re-scored. Free account.
Alert meAnalysisAIv1 – Sep 30, 2026
Out-of-bounds write in Android's AudioDecoder (audio_decoder.cc, AudioDecoder::HandleProduceRequest) lets a remote, unauthenticated attacker execute code by feeding crafted audio data into the decoder path, with no user interaction required, as the faulty bounds check on the produce request's size/offset fields can be bypassed. CVSS 3.1 is 9.8 (AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H) with CWE-787; the practical prerequisite is that some application or service must route untrusted audio into this decoder and the device must be running an unpatched build. There is no public exploit identified at time of analysis, and the only vendor reference is the December 2025 Pixel bulletin, so real-world exposure appears limited to Pixel/AOSP builds carrying this decoder below the 2025-12-01 patch level.
RemediationAI
Vendor-released patch: the fix is delivered in the December 2025 Pixel/Android security update, patch level 2025-12-01, per https://source.android.com/security/bulletin/pixel/2025-12-01; Pixel users should install the OTA immediately and verify under Settings > About phone > Android version > Android security update that the level reads 2025-12-01 or later, and enterprise fleets should treat that patch level as the compliance baseline for affected devices. If patching cannot be applied immediately, compensating controls are limited because the decoder is a platform component: restrict which applications and services may process audio from untrusted network sources (for example, disable auto-download and auto-play of media in messaging and social apps, and block or quarantine untrusted audio attachments at the email/web gateway), and consider disabling or removing non-essential apps that accept arbitrary audio input - the trade-off being reduced functionality and a real risk that a system service or a preferred app still routes untrusted audio into the same decoder, leaving the exposure only partially mitigated. Because the vulnerable code runs in an isolated media codec process, applying OS-level hardening and keeping other components current does not eliminate this bug but may reduce the value of code execution obtained there; the only complete remediation is installing the 2025-12-01 (or later) security patch level.
Affects your stack
Find out whether the stacks and servers you run are on an affected version.
Checking this CVE against your own stacks and servers is part of the Basic plan and up.
Technical ContextAI · show details
The flaw is a classic CWE-787 out-of-bounds write rooted in an incorrect bounds check inside AudioDecoder::HandleProduceRequest. The affected component is part of Android's native audio/media decoding stack, where a producer submits a 'produce request' whose size and offset fields describe a destination buffer region; the vulnerable check fails to reject a crafted size/offset pair, so the decoder writes past the intended buffer in native code. Because the CPE identifies the platform as cpe:2.3:o:google:android (Google Android, all versions unqualified), this is an AOSP-level implementation issue rather than a third-party app bug, and the fix is distributed through the monthly Android/Pixel security update program rather than an app store update. The CVSS vector AV:N/AC:L/PR:N/UI:N indicates the attack is network-reachable, low complexity, unauthenticated, and needs no victim interaction, consistent with a zero-click crafted-media delivery model in which the audio payload is decoded automatically by a system or privileged service. The decode operation typically occurs inside an isolated media codec process, so the immediate impact is code execution in that sandboxed decoder context, which an attacker would then need to chain with a separate sandbox-escape to reach higher privileges.
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:N/S:U/C:H/I:H/A:H
Unauthenticated, no-interaction OOB write reachable via remotely delivered audio giving code execution, so AV:N/PR:N/UI:N with full C/I/A impact.
Primary rating from NVD.
CVSS VectorNVD
DescriptionNVD
In AudioDecoder::HandleProduceRequest of audio_decoder.cc, there is a possible out of bounds write due to an incorrect bounds check. This could lead to remote code execution with no additional execution privileges needed. User interaction is not needed for exploitation.
Executive ViewAI
A critical memory-safety flaw (CVE-2025-36937) in Android's built-in audio decoder allows an unauthenticated, remote attacker to run code on a device simply by getting crafted audio into a decoding path, with no user interaction required. Exposure is currently limited to Pixel and AOSP-based devices running builds below the 2025-12-01 patch level, but those devices cannot be fully remediated today because no vendor-released patch has been identified at time of analysis. Leadership should care because mobile devices are the primary work platform for remote and executive staff, and a compromised handset can expose corporate email, credentials, and network access without any user mistake to detect.
No vendor-released patch identified at time of analysis. Interim mitigations: reduce attack surface by removing or blocking applications and services that accept externally sourced audio for decoding; enforce app sandboxing and least-privilege mobile policies; restrict mobile endpoint access to sensitive resources via conditional access and network segmentation; monitor for repeated audio decoder crashes as an exploitation indicator; prioritize replacement or isolation of unpatched devices for high-risk users until an Android patch level of 2025-12-01 or later is confirmed. Note: the only vendor reference is the December 2025 Pixel security bulletin, and the vulnerability description states no public exploit identified at time of analysis.
Threat Intelligence Overview
NVD
CVE.org
EPSS
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
A buffer overflow occurs when a program writes more data to a memory buffer than it was allocated to hold, causing the excess data to spill into adjacent memory regions.
Memory corruption occurs when a program writes data beyond the boundaries of allocated memory regions or accesses memory in unintended ways, violating the integrity of the process's address space.
Remote Code Execution represents the critical moment when an attacker successfully runs arbitrary code on a target system without physical access.
GitHub PoC
Exploit-DB