Attack ChainAIDerived
Hypothetical attack flow derived from CVE metadata – not a confirmed exploit chain
-
AccessAccess multi-tenant Pillow Arrow endpoint
-
DeliveryRequest Arrow export of small/cropped image
-
ExploitLibrary exports from blocks[0].ptr ignoring alignment padding
-
ExecutionBuffer returns stale heap/other users' pixels
-
ImpactDisclose cross-tenant memory
Vulnerability Assessment
| Weakness |
CWE
ASVS 5.0
V14 - Data Protection Sensitive info in resource not removed before reuse
|
| Exploitation Conditions | Exploitation requires a specific non-default deployment: the application must configure a non-default PILLOW_ALIGNMENT (so aligned_linesize adds row padding or an aligned offset before the first row) AND export images to the Arrow format via the __arrow_c_array__ / Arrow C interface. The leak is most pronounced for small or cropped images where per-row padding is significant, and for the information to cross trust boundaries the deployment must be multi-user (one user's export must be able to reach another user). Limiting factors: default Pillow configurations do not set custom alignment and are not affected; single-user or non-Arrow workflows cannot leak cross-tenant data; affects only versions 11.2.1 through 12.x, fixed in 13.0.0. The AT:P metric reflects that these configuration prerequisites are outside the attacker's control. |
| Impact | This is a genuine but narrowly-scoped information-disclosure flaw (CWE-226) that should be treated as a moderate, not top, priority. The only severity signal provided is a vendor/GitHub CVSS 4.0 vector (CVSS:4.0/AV:N/AC:L/AT:P/PR:N/UI:N/VC:L/VI:N/VA:N, score 6.3): network vector, no privileges or user interaction, but with AT:P (a present attack requirement) and only Low confidentiality impact and no integrity/availability impact. The AT:P flag is the key nuance - exploitation is conditional on the application enabling PILLOW_ALIGNMENT and exporting images to Arrow, so this is not a straightforward remote attack despite AV:N. There is no EPSS score, no CISA KEV listing, and no public exploit referenced in the data, so exploitation status is 'no public exploit identified at time of analysis'. No SSVC decision data was supplied. The disclosed content is uninitialized padding or stale heap from previously freed images, which can leak another user's pixels or process memory - meaningful in multi-tenant image-processing services but useless against single-user deployments. Net: a real, fix-worthy bug for multi-user services that use Pillow's Arrow export with custom alignment, but low urgency for the large majority of Pillow users who never touch alignment or Arrow export. |
| Exploit ScenarioDerived | A multi-tenant image service built on Pillow is configured with a non-default PILLOW_ALIGNMENT and exposes image data to clients via the Arrow C array interface. When one user requests an Arrow export of a small or cropped image, the buffer sent back includes padding bytes or stale heap left over from a previously freed image belonging to another user, leaking their pixels or other process memory. No public proof-of-concept was referenced, though the fixing PR (#10172) includes regression tests that demonstrate the non-contiguous/aligned conditions. |
Get an e-mail when this CVE is exploited, patched or re-scored. Free account.
Alert meAnalysisAIv1 – Oct 21, 2026
Pillow 11.2.1 through 12.x can disclose uninitialized padding, bytes preceding the first row, or stale heap contents when an application sets a non-default PILLOW_ALIGNMENT and exports images through the Arrow C interface (__arrow_c_array__/__arrow_c_schema__). The flaw is limited to deployments that both customize alignment and use Arrow export, with meaningful cross-user exposure requiring a multi-user context where one user's exported buffer can reach another user; default Pillow configurations are not affected. No public exploit code or CISA KEV entry was identified at time of analysis, and the vendor released Pillow 13.0.0 as the fix.
RemediationAI
Upgrade to Pillow 13.0.0, the vendor-released fixed version; see the release at https://github.com/python-pillow/Pillow/releases/tag/13.0.0, advisory https://github.com/python-pillow/Pillow/security/advisories/GHSA-654x-cwwv-5j9c, fix PR https://github.com/python-pillow/Pillow/pull/10172, and commit https://github.com/python-pillow/Pillow/commit/19f64946eb247b973dd4560be7dea3bf979d2e6a. If immediate upgrade is not possible, remove non-default PILLOW_ALIGNMENT settings and disable or block Arrow export paths (__arrow_c_array__/__arrow_c_schema__) for untrusted or multi-tenant images; reverting alignment to the default gives up any alignment-related performance or ABI benefit, while blocking Arrow export breaks Arrow interoperability. Deployments that cannot drop alignment should isolate Arrow export to a single-user process or copy and scrub exported buffers before sharing, but those are compensating controls with residual risk and do not replace upgrading to 13.0.0.
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
Pillow stores pixel data in C-level image blocks and exposes an Arrow C data interface via __arrow_c_array__/__arrow_c_schema__. The vulnerable export path assumes rows are tightly contiguous and starts the Arrow buffer at im->blocks[0].ptr; when PILLOW_ALIGNMENT causes aligned_linesize to include per-row padding or an aligned offset before the first row, the computed buffer can span padding, pre-row bytes, or freed-image heap contents instead of only the intended pixels. This is CWE-226 (Sensitive Information in Resource Not Removed Before Reuse): stale or uninitialized memory is exposed when the image resource is reused for Arrow export. The independent assessment's CVSS 3.1 vector AV:N/AC:H/PR:N/UI:N/S:U/C:L/I:N/A:N reflects that exploitation needs the non-default alignment setting plus Arrow export, and cross-tenant impact needs a multi-user deployment; the CPE cpe:2.3:a:python-pillow:pillow:*:*:*:*:*:*:*:* identifies Pillow as the affected package, with the vendor advisory narrowing affected versions to 11.2.1 through 12.x.
Know when Pillow gets its next vulnerability
- 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
CVSS:4.0/AV:N/AC:L/AT:P/PR:N/UI:N/VC:L/VI:N/VA:N/SC:N/SI:N/SA:N
AC:H because exploitation depends on non-default alignment plus Arrow export and on heap/timing to recover stale data; C:L only, with no integrity or availability impact.
Primary rating from Vendor (GitHub_M).
CVSS VectorVendor: GitHub_M
Lifecycle Timeline
2DescriptionCVE.org
Pillow is a Python imaging library. From 11.2.1 until 13.0.0, Arrow export assumes that an image's rows are tightly contiguous and exports from im->blocks[0].ptr even when PILLOW_ALIGNMENT makes aligned_linesize include row padding or an aligned offset before the first row. The resulting Arrow buffer can include uninitialized padding, bytes preceding the first row, or stale heap contents from previously freed images. Applications that configure alignment and export images to Arrow can therefore disclose one user's pixels or other process memory to another user. This issue is fixed in version 13.0.0.
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
Information disclosure occurs when an application unintentionally exposes sensitive data that aids attackers in reconnaissance or directly compromises security.
libImaging/TgaRleDecode.c in Pillow 9.1.0 has a heap buffer overflow in the processing of invalid TGA image files. Rated
Denial of service in Python Pillow (versions 8.2.0 through 12.2.0) allows remote attackers to exhaust memory during JPEG
Out-of-bounds memory disclosure in Python Pillow before 12.3.0 lets an attacker who supplies a crafted McIdas AREA image
Heap out-of-bounds write in Python Pillow prior to 12.3.0 lets an attacker who controls the rank-filter size parameter c
Uncontrolled memory consumption in Python Pillow (PIL) versions 5.1.0 through 12.2.x allows a remote attacker to cause d
The package pillow 5.2.0 and before 8.3.2 are vulnerable to Regular Expression Denial of Service (ReDoS) via the getrgb
Pillow is a Python imaging library. In versions 11.2.0 to before 11.3.0, there is a heap buffer overflow when writing a
Integer overflow in the ImagingResampleHorizontal function in libImaging/Resample.c in Pillow before 3.1.1 allows remote
Python Image Library (PIL) 1.1.7 and earlier and Pillow 2.3 might allow remote attackers to execute arbitrary commands v
Pillow versions 11.0.0 through 12.x (fixed in 13.0.0) can reuse a stale native ImagingCore buffer when a multi-frame ima
Pillow through 8.2.0 and PIL (aka Python Imaging Library) through 1.1.7 allow an attacker to pass controlled parameters
An issue was discovered in Pillow before 8.1.1. Rated critical severity (CVSS 9.8), this vulnerability is remotely explo
EUVD-2026-107575
NVD
CVE.org
EPSS
GitHub
VulDB
Exploit-DB