CVE-2026-88053

A flaw was found in Tesseract, an open-source Optical Character Recognition (OCR) engine. A local attacker could provide a specially crafted .traineddata file, which, when processed, could lead to a heap out-of-bounds write. This vulnerability occurs due to unvalidated counts used as loop bounds during legacy-classifier initialization. Successful exploitation could result in heap corruption, causing a crash or potentially controlled corruption of memory.

This vulnerability is rated as an Important severity flaw because an adversary who successfully triggers the memory corruption can execute arbitrary code within the privileges of the running application, rather than solely causing a denial of service. In Red Hat Enterprise Linux, Tesseract operates as an optical character recognition utility and shared library that typically handles language models bundled directly from verified system packages. Exploitation requires an application to load a malicious training model, meaning risk is concentrated in automated pipelines or services configured to accept external, unvetted language data files.

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 vulnerability, ensure that Tesseract only processes .traineddata files from trusted sources. Avoid loading or processing .traineddata files from untrusted or unverified origins, as these could be specially crafted to trigger the heap corruption. If Tesseract must process untrusted input, consider running the application in a sandboxed environment to limit the potential impact of successful exploitation.