Database/AI/ML frameworks & serving
Keras: Safe-mode bypass in Keras 3.0.0–3.10.0
CVSS 7.8CVE-2025-8747AI/ML frameworks & servingcurated
Impact
Safe-mode bypass in Keras 3.0.0–3.10.0 → arbitrary code execution
Who can reach it
Customer-supplied .keras file
What to do
Upgrade past 3.10.0; bypass of the CVE-2025-1550 fix
Fleet impact
How widespread
Very common - the bypass of the 3.9.0 fix, so operators who patched once are still exposed
Cost to remediate
**Image rebuild** again - the second round, which is exactly the fleet-wide-patch fatigue pattern
Why it hits the whole fleet
Reuse of internal functionality re-enables arbitrary code execution on load_model, proving the model-file-as-code class is not closable by a single patch
References
Related entries
- Keras: Code execution from crafted `.keras` archive despite safe modeCVE-2025-9906 · KerasHigh
- Keras: Deserialization of untrusted data in 3.11.0–3.11.2CVE-2025-49655 · KerasCritical
- NVIDIA Megatron-LM: unsafe deserialization of untrusted data allows code executionCVE-2026-100967 · NVIDIA Megatron-LMHigh
- NVIDIA Megatron-LM: unsafe deserialization in dataset loading allows code executionCVE-2026-100968 · NVIDIA Megatron-LM (dataset loading)High
- MLflow: pickle evaluation artifact deserialization runs arbitrary code when a model is loadedCVE-2026-118623 · MLflow model artifact loading (PickleEvaluationArtifact)High
- MLflow: loading a SHAP explainer model deserializes untrusted data and runs arbitrary codeCVE-2026-118627 · MLflow SHAP explainer loading (load_explainer)High
This entry is curated: imported from vendor advisories with machine assistance, not yet individually verified. Confirm against your vendor's advisory before acting, and report anything wrong.