{"slug": "black-box-red-teaming-of-agentic-ai-a-taxonomy-driven-framework-for-automated", "title": "Black-Box Red Teaming of Agentic AI: A Taxonomy-Driven Framework for Automated Risk Discovery", "summary": "A new arXiv paper (2609.09647v1) presents a black-box red teaming framework that found 56.25% average governance risk, 65% privacy risk in multi-agent configurations, and agent behavior vulnerabilities reaching 85% across two agent architectures (CrewAI and AutoGen) and four base models. The framework combines a seven-domain taxonomy, the fully automated SAGE-RT red teaming method generating 120 adversarial scenarios per domain, and human-validated evaluation using LLM judges. The authors report the approach identifies critical architectural vulnerabilities without privileged access, offering a scalable path toward safer agent deployments.", "body_md": "arXiv:2609.09647v1 Announce Type: new \nAbstract: Agentic systems are rapidly moving to production, where they read untrusted inputs, call tools with real permissions, and act autonomously, expanding the security surface beyond chat-only models. Yet standard evaluations remain single-turn and fail to capture multi-step agent vulnerabilities. We present a systematic black-box framework for risk-aware agent evaluation requiring only basic system descriptions. Our approach introduces: (1) a seven-domain taxonomy mapping observable behaviors to risk categories, (2) fully automated SAGE-RT red teaming producing 120 adversarial scenarios per domain, and (3) human-validated evaluation using LLM judges. Empirical validation across two agent architectures (CrewAI and AutoGen) with four base models reveals alarming patterns: 56.25\\% average governance risk, 65\\% privacy risk in multi-agent configurations, and agent behavior vulnerabilities reaching 85\\%. Our black-box approach effectively identifies critical architectural vulnerabilities without privileged access, providing a scalable path toward safer agent deployments.", "url": "https://wpnews.pro/news/black-box-red-teaming-of-agentic-ai-a-taxonomy-driven-framework-for-automated", "canonical_source": "https://arxiv.org/abs/2609.09647", "published_at": "2026-09-11 04:00:00+00:00", "updated_at": "2026-09-11 04:27:57.424613+00:00", "lang": "en", "topics": ["ai-safety", "ai-agents", "ai-research", "ai-ethics"], "entities": ["CrewAI", "AutoGen", "SAGE-RT", "arXiv"], "alternates": {"html": "https://wpnews.pro/news/black-box-red-teaming-of-agentic-ai-a-taxonomy-driven-framework-for-automated", "markdown": "https://wpnews.pro/news/black-box-red-teaming-of-agentic-ai-a-taxonomy-driven-framework-for-automated.md", "text": "https://wpnews.pro/news/black-box-red-teaming-of-agentic-ai-a-taxonomy-driven-framework-for-automated.txt", "jsonld": "https://wpnews.pro/news/black-box-red-teaming-of-agentic-ai-a-taxonomy-driven-framework-for-automated.jsonld"}}