8.8Alta

CVE-2026-1462Safe Mode Bypass in keras-team/keras

keras-team keras-team/keras; Red Hat Red Hat OpenShift AI 2.25; Red Hat Red Hat OpenShift AI 3.3; Red Hat Red Hat OpenShift AI (RHOAI) Publicada: (hace 7 meses)

Descripción

A vulnerability in the `TFSMLayer` class of the `keras` package, version 3.13.0, allows attacker-controlled TensorFlow SavedModels to be loaded during deserialization of `.keras` models, even when `safe_mode=True`. This bypasses the security guarantees of `safe_mode` and enables arbitrary attacker-controlled code execution during model inference under the victim's privileges. The issue arises due to the unconditional loading of external SavedModels, serialization of attacker-controlled file paths, and the lack of validation in the `from_config()` method.

Debilidad (CWE): CWE-502

Datos: CVE.org, NVD, CISA KEV, FIRST EPSS, Ubuntu. Se publican tal como los indican las fuentes, en inglés.

Productos y versiones afectados

FabricanteProductoVersiones
keras keras 3.13.0
keras-team keras-team/keras unspecified – < 3.13.2
openshift ai 2.25 —
openshift ai 3.3 —
redhat openshift ai 2.25 – < 2.25.7
Red Hat Red Hat OpenShift AI 2.25 all versions
Red Hat Red Hat OpenShift AI 3.3 all versions
Red Hat Red Hat OpenShift AI (RHOAI) all versions

Referencias y parches

Coincidencias CPE (NVD) (2)

  • cpe:2.3:a:keras:keras:3.13.0:*:*:*:*:*:*:*
  • cpe:2.3:a:redhat:openshift_ai:*:*:*:*:*:*:*:* ≥ 2.25 < 2.25.7