Impact
Pillow’s EXIF parsing routine processes the PNG eXIf chunk in a way that causes a quadratic‑time copy operation for each repeated Exif header. A crafted PNG containing many header repeats forces the library to copy shrinking chunks repeatedly, resulting in excessive CPU usage and a denial of service. The vulnerability does not provide direct code execution or privilege escalation; its impact is limited to disrupting service availability.
Affected Systems
The flaw affects Python Pillow versions 11.0.0 through 12.x.x, inclusive. All products that depend on these releases for image processing, such as web applications that upload or manipulate PNG files, are potentially impacted.
Risk and Exploitability
The CVSS score of 7.1 reflects a high impact, while the EPSS score of 0.00333 (0.333%) indicates a very low likelihood of exploitation. The flaw is not listed in the CISA KEV catalog. Attackers can exploit the weakness by supplying a malicious PNG file—either via a local file upload, a remote web endpoint, or any automated image ingestion process that uses Pillow’s EXIF handling. The likely attack vector is through application‑level image upload workflows where end users submit pictures.
OpenCVE Enrichment
Github GHSA