On March 30, according to the latest news from the GitHub community, the Ant AI Security Lab recently conducted a specialized security audit on the open-source autonomous agent framework OpenClaw. During a three-day in-depth inspection, the team submitted a total of 33 security vulnerability reports. Among them, OpenClaw has confirmed and fixed 8 vulnerabilities in its latest released version 2026.3.28, including 1 critical-level vulnerability, 4 high-risk vulnerabilities, and 3 medium-risk vulnerabilities. Ant Group stated that it will continue to monitor the security risks of OpenClaw and provide support for the safe and stable application of AI agents in the industry.
Ant AI Security Lab Discovers Multiple High-Risk Vulnerabilities in OpenClaw and Assists in Rapid Fixing
15.0K views
Related
6000 Reddit Comments Reveal the Truth About AI Programming Assistant Security: Cursor Has the Most Incidents, Excessive Permissions Are the Biggest Risk
15.2K
AI-Booked Fitness Classes Turn into Hacking: Taking Over the First Place in Line by Yourself - Australia's First Intelligent Agent Cyber Attack
12.7K
Ant Group Open Sources Avernet: Solving the Challenges of Multi-Agent Collaboration
32.7K
OpenAI AI Agent Breaks Through Sandboxing and Infiltrates Open-Source Platform Hugging Face, AI Security Boundary Compromised
15.0K
Peking University's Talented Female Scholar Wang Li Turns Again: Resigns from AI Entrepreneurship Due to Severe Physical Expenditure, 'People-Oriented' Strikes Industry Pain Points
16.9K
New Breakthrough in Embodied Intelligence: Ant Group Open Sources LingBot-Vision, Enabling Robots to Have a Sense of Space
15.4K