LLM Security: Vulnerabilities, Attacks, Defenses, and Countermeasures

As large language models LLMs continue to evolve, it is critical to assess the security threats and vulnerabilities that may arise both during their training phase and after models have been deployed. This survey seeks to define and categorize the various attacks targeting LLMs, distinguishing...

Picklescan Vulnerabilities Could Let Hackers Bypass AI Security Checks

Sonatype researchers uncover critical vulnerabilities in picklescan. Learn how these flaws impact AI model security, Hugging Face, and…...

CVE-2025-1945 picklescan - Zip Flag Bit Exploit Crashes Picklescan But Not PyTorch

picklescan before 0.0.23 fails to detect malicious pickle files inside PyTorch model archives when certain ZIP file flag bits are modified. By flipping specific bits in the ZIP file headers, an attacker can embed malicious pickle files that remain undetected by PickleScan while still being...

BIT-NODE-MIN-2024-21896

The permission model protects itself against path traversal attacks by calling path.resolve on any paths given by the user. If the path is to be treated as a Buffer, the implementation uses Buffer.from to obtain a Buffer from the result of path.resolve. By monkey-patching Buffer internals, namely...

Securely Build AI/ML Applications in the Cloud with Rapid7 InsightCloudSec

It’s been little over a year since ChatGPT was released, and oh how much has changed. Advancements in Artificial Intelligence and Machine Learning have marked a transformative era, influencing virtually every facet of our lives. These innovative technologies have reshaped the landscape of natural...

Google TensorFlow 缓冲区错误漏洞

Google TensorFlow is an end-to-end open source platform for machine learning from Google Google. Google Tensorflow is vulnerable to a buffer overflow vulnerability that could be exploited by an attacker to build a TFLite model that allows limited reads and writes to the outside of arrays in TFLit...