Sign in Subscribe

Topic

AI security

A collection of 5 issues

OpenClaw Meltdown Exposes Critical Vulnerabilities in AI Agent Ecosystems

The OpenClaw Meltdown exposed nine CVEs and 2,200 malicious skills within AI agent ecosystems, marking the largest real-world test of the OWASP Agentic Top 10. This incident highlights critical vulnerabilities and underscores the urgent need for robust security measures in AI systems. The event

OpenClaw Meltdown Exposes 9 CVEs, 2,200 Malicious Skills

The OpenClaw Meltdown exposed nine CVEs and 2,200 malicious skills, highlighting critical vulnerabilities in AI agent ecosystems. This incident tested the OWASP Agentic Top 10 and underscores the need for improved security measures in AI systems. The discovery emphasizes the risks posed by mali

New Defense Thwarts Attacks on AI Language Agents

ICON, a new defense mechanism, effectively neutralizes Indirect Prompt Injection (IPI) attacks on Large Language Model (LLM) agents. By leveraging a probing-to-mitigation framework, ICON achieves a competitive 0.4% Attack Success Rate (ASR) and a significant 50% task utility gain. This ensures

New Defense Mechanism Shields AI Agents from Prompt Injection Attacks

ICON, a new defense mechanism, protects Large Language Model (LLM) agents from Indirect Prompt Injection (IPI) attacks. The probing-to-mitigation framework detects and neutralizes malicious instructions embedded in retrieved content, improving task utility by over 50% and achieving a low 0.4% A

TrojAI Report Exposes AI System Vulnerabilities

A final report from IARPA's TrojAI program exposes vulnerabilities in AI systems, specifically AI Trojans, which are malicious backdoors embedded within AI models that can cause failures or hijacking. The report emphasizes the need for robust defenses against adversarial attacks to safeguard AI