CVE-2026-88051

A flaw was found in Tesseract, an open-source Optical Character Recognition (OCR) engine. A heap out-of-bounds write vulnerability exists in the GenericVector::read function when processing a specially crafted .traineddata model. By manipulating the reserved and size_used_ fields, a local attacker can cause heap corruption, a crash, or potentially controlled corruption, which could lead to arbitrary code execution.

This vulnerability is rated as Important for Red Hat Enterprise Linux because successful exploitation of the memory corruption flaw can result in arbitrary code execution within the context of the user or process running the OCR engine. In standard Red Hat deployments, Tesseract relies on verified system-provided language packs, limiting exposure unless an application or user is tricked into loading an untrusted, malicious trained data file. Because triggering the condition requires local file interaction or supplying a specially prepared model to an automated ingestion pipeline, remote unauthenticated exploitation without user interaction is not possible.

The CVSS score(s) listed for this vulnerability do not reflect the associated product's status, and are included for informational purposes to better understand the severity of this vulnerability.

To mitigate this issue, avoid processing .traineddata model files from untrusted or unknown sources. Ensure that Tesseract only processes models from trusted origins. If Tesseract is not required, consider removing the tesseract package to eliminate the attack surface.