AI frameworks, including Meta’s Llama, are prone to automatic Python deserialization by pickle that could lead to remote code execution. Meta’s large language model (LLM) framework, Llama, suffers a ...

Thirteen critical vulnerabilities have been found in the vm2 JavaScript sandbox package that could allow an attacker’s code to escape the container and do nasty things to IT environments. As a result, ...

Lambda Layers in third party TensorFlow-based Keras models allow attackers to inject arbitrary code into versions built prior to Keras 2.13 that may then unsafely run with the same permissions as the ...

A vulnerability chain dubbed AutoJack in Microsoft's AutoGen Studio interface for prototyping AI agents could let attackers ...

A new vulnerability has been discovered in the R programming language that allows arbitrary code execution upon deserializing specially crafted RDS and RDX files. R is an open-source programming ...