AI Tool Discovers High-Severity ReDoS Vulnerability in NLTK

sachdh · x · 2026-08-28

Kira, an AI tool, discovered a high-severity ReDoS (Regular Expression Denial of Service) vulnerability in NLTK, a Python library with over 1.5 billion downloads and 25 years of history. The tool not only identified the flaw but also autonomously verified that it was exploitable. NLTK sits deep in the Python dependency tree, underpinning the stack of many AI-native applications.

Original post →

More from Safety

Safety channel →