EUVD-2026-45872

A vulnerability in keras-team/keras version 3.15.0 allows unsafe deserialization of attacker-controlled PyTorch pickle data through the public keras.layers.TorchModuleWrapper.fromconfig method. This method invokes torch.load..., weightsonly=False without requiring an explicit unsafe opt-in, such ...

CVE-2026-12481

A vulnerability in keras-team/keras version 3.14.0 allows for arbitrary code execution due to improper handling of deserialization in the Lambda layer. Specifically, the raiseforlambdadeserialization function fails to enforce the safe-mode guard when safemode is set to None, which is the default...

CVE-2026-12481

The CVE affects keras-team/keras 3.14.0, where deserialization in the Lambda layer can bypass the safe-mode guard. The root cause is _raise_for_lambda_deserialization() treating safe_mode=None as unset/deny, allowing attacker-controlled marshal bytecode to be deserialized. Affected call sites inc...

CVE-2026-12481 Deserialization of Untrusted Data in keras-team/keras

A vulnerability in keras-team/keras version 3.14.0 allows for arbitrary code execution due to improper handling of deserialization in the Lambda layer. Specifically, the raiseforlambdadeserialization function fails to enforce the safe-mode guard when safemode is set to None, which is the default...

Security Bulletin: Multiple security vulnerabilities in Python affect IBM Robotic Process Automation

Summary Multiple security vulnerabilities in Python affect IBM Robotic Process Automation. Python is used by IBM Robotic Process Automation as part of Watson NLP. This bulletin identifies the fixes to resolve the vulnerabilities. Vulnerability Details CVEID:CVE-2025-8747 DESCRIPTION: A safe mode...

Linux Distros Unpatched Vulnerability : CVE-2025-49655

The Linux/Unix host has one or more packages installed that are impacted by a vulnerability without a vendor supplied patch available. - Deserialization of untrusted data can occur in versions of the Keras framework running versions 3.11.0 up to but not including 3.11.3, enabling a maliciously...

GHSA-CVHH-Q5G5-QPRP Keras framework vulnerable to deserialization of untrusted data

Deserialization of untrusted data can occur in versions of the Keras framework running versions 3.11.0 up to but not including 3.11.3, enabling a maliciously uploaded Keras file containing a TorchModuleWrapper class to run arbitrary code on an end user’s system when loaded despite safe mode being...

EUVD-2025-34892

Keras framework vulnerable to deserialization of untrusted data...

Keras framework vulnerable to deserialization of untrusted data

Deserialization of untrusted data can occur in versions of the Keras framework running versions 3.11.0 up to but not including 3.11.3, enabling a maliciously uploaded Keras file containing a TorchModuleWrapper class to run arbitrary code on an end user’s system when loaded despite safe mode being...

CVE-2025-49655

Deserialization of untrusted data can occur in versions of the Keras framework running versions 3.11.0 up to but not including 3.11.3, enabling a maliciously uploaded Keras file containing a TorchModuleWrapper class to run arbitrary code on an end user’s system when loaded despite safe mode being...

CVE-2025-49655

CVE-2025-49655 affects the Keras framework with deserialization of untrusted data in versions 3.11.0 up to, but not including, 3.11.3. A maliciously uploaded Keras file containing a TorchModuleWrapper class can execute arbitrary code on an end user’s system when loaded, even if safe mode is enabl...

GHSA-36FQ-JGMW-4R9C Keras is vulnerable to Deserialization of Untrusted Data

Arbitrary Code Execution in Keras Keras versions prior to 3.11.0 allow for arbitrary code execution when loading a crafted .keras model archive, even when safemode=True. The issue arises because the archive’s config.json is parsed before layer deserialization. This can invoke...

📄 Keras 2.15 Remote Code Execution

This exploit abuses insecure deserialization in Keras model loading. By embedding a malicious "function" object inside a .keras file or config.json, an attacker can execute arbitrary system commands as soon as the model is loaded using keras.models.loadmodel or modelfromjson. This proof of concep...