{"slug": "trustworthy-agentic-ai-a-comprehensive-cybersecurity-and-systems-survey-on-and", "title": "Trustworthy Agentic AI: A Comprehensive Cybersecurity and Systems Survey on Threat Landscapes, Defense Architectures, and Open Challenges", "summary": "A new arXiv survey (2609.13731v1) synthesizes 206 foundational studies and regulatory standards into a systems-security reference framework for trustworthy agentic AI, formalizing the general agent architecture as a stateful 5-tuple and establishing a 6-dimensional trustworthiness taxonomy covering security, safety, privacy, explainability, fairness, and accountability. The survey analyzes threat surfaces across intra-execution loops and interaction planes and proposes a multi-layered zero-trust defense-in-depth architecture integrating Dual-LLM isolation, Capability-Based Access Control, kernel eBPF probes, and sandboxed runtimes. It warns that granting probabilistic neural cores execution authority across filesystems, networks, and cloud infrastructure dissolves classical security perimeters, exposing a Turing-complete blast radius where untrusted data represents executable instructions.", "body_md": "arXiv:2609.13731v1 Announce Type: new \nAbstract: The transition from passive foundation models to autonomous, goal-directed agentic AI systems has introduced unprecedented capabilities by coupling recursive cognitive reasoning loops, persistent memory architectures, live tool execution planes, and multi-agent collaboration topologies. However, granting probabilistic neural cores execution authority across filesystems, networks, and cloud infrastructure dissolves classical security perimeters: natural language simultaneously serves as input data, internal control code, and communication protocols, exposing a Turing-complete blast radius where untrusted data represents executable instructions. This survey delivers a comprehensive systems-security reference framework for trustworthy agentic AI, synthesizing 206 foundational studies and regulatory standards. We formalize the general agent architecture as a stateful 5-tuple and establish a 6-dimensional trustworthiness taxonomy covering security, safety, privacy, explainability, fairness, and accountability. We systematically analyze threat surfaces across intra-execution loops and interaction planes, formulate a multi-layered zero-trust defense-in-depth architecture integrating Dual-LLM isolation, Capability-Based Access Control, kernel eBPF probes, and sandboxed runtimes, review standardized evaluation benchmarks, and map technical controls to international AI governance frameworks.", "url": "https://wpnews.pro/news/trustworthy-agentic-ai-a-comprehensive-cybersecurity-and-systems-survey-on-and", "canonical_source": "https://www.machinebrief.com/news/trustworthy-agentic-ai-a-comprehensive-cybersecurity-and-sys-bgyo", "published_at": "2026-09-15 04:00:00+00:00", "updated_at": "2026-09-15 04:33:33.877832+00:00", "lang": "en", "topics": ["ai-safety", "ai-agents", "ai-research", "ai-policy", "artificial-intelligence"], "entities": ["arXiv", "Dual-LLM isolation", "Capability-Based Access Control", "eBPF"], "alternates": {"html": "https://wpnews.pro/news/trustworthy-agentic-ai-a-comprehensive-cybersecurity-and-systems-survey-on-and", "markdown": "https://wpnews.pro/news/trustworthy-agentic-ai-a-comprehensive-cybersecurity-and-systems-survey-on-and.md", "text": "https://wpnews.pro/news/trustworthy-agentic-ai-a-comprehensive-cybersecurity-and-systems-survey-on-and.txt", "jsonld": "https://wpnews.pro/news/trustworthy-agentic-ai-a-comprehensive-cybersecurity-and-systems-survey-on-and.jsonld"}}