Detection Confidence as the Authorization Gate: Integrating the Detection Lifecycle Codex and Agentic Security Governance Framework into a Closed-Loop Autonomous SOC Architecture
DOI:
https://doi.org/10.63282/3050-9262.IJAIDSML-V7I3P111Keywords:
Autonomous SOC, Detection Confidence, Security Orchestration, Agentic AI Security, Detection Engineering, Authorization Policy, Human-In-The-Loop, Closed-Loop Response, Telemetry Integrity, Explainable Automation, Security Governance, Adaptive Incident ResponseAbstract
Autonomous response should not be triggered merely because a detection fires. This paper proposes detection confidence as an authorization gate that determines whether a security action may execute automatically, requires human review, or must be held. Two frameworks are specified to make that gate operable. The Detection Lifecycle Codex governs a detection from use-case definition through telemetry contract, adversarial validation, confidence calibration, authorization classification, runtime assurance, and retirement, and it assigns each detection an authorization class rather than a severity label. The Agentic Security Governance Framework governs the security agents that recommend or execute response, controlling identity, tool permissions, action boundaries, approval requirements, and rollback. A closed-loop architecture integrates the two: evidence is assembled, confidence is assessed across six independent dimensions, action risk is classified separately, and a policy decision point outside the agent's trust boundary arbitrates between them before anything executes. Outcomes are recorded and returned to confidence calibration and authorization policy. The arbitration rule is asymmetric by design, in that confidence can only authorize consideration of an action and never overrides the action-risk class, so no confidence score unlocks an irreversible operation. The architecture is specified and evaluated analytically. It is not reported as a measured production deployment, and the paper states which quantities remain unmeasured and what a controlled pilot would have to produce.
References
[1] B. A. Alahmadi, L. Axon, and I. Martinovic, "99% False Positives: A Qualitative Study of SOC Analysts' Perspectives on Security Alarms," in Proc. 31st USENIX Security Symposium, 2022, pp. 2783-2800.
[2] A. Mohsin, H. Janicke, A. Ibrahim, I. H. Sarker, and S. Camtepe, "A Unified Framework for Human AI Collaboration in Security Operations Centers with Trusted Autonomy," arXiv:2505.23397, 2025.
[3] J. Roy and S. K. Singh, "AgentSOC: A Multi-Layer Agentic AI Framework for Security Operations Automation," arXiv:2604.20134, 2026.
[4] M. H. Saju and A. Azim, "Toward Autonomous SOC Operations: End-to-End LLM Framework for Threat Detection, Query Generation, and Resolution in Security Operations," arXiv:2604.27321, 2026.
[5] R. Sahay, B. Eapen, W. Meng, M. R. Al Mamun, N. K. Dora, M. Sumasadan, S. K. Tetarave, and E. De La Cruz, "Policy-Guided Threat Hunting: An LLM enabled Framework with Splunk SOC Triage," arXiv:2603.23966, 2026.
[6] X. Cadet, A. V. Singh, H. Mamania, E. Koh, A. Fitts, D. Van Bruggen, S. Boboila, P. Chin, and A. Oprea, "Retrieval-Augmented LLMs for Security Incident Analysis," arXiv:2603.18196, 2026.
[7] S. J. Lazer, K. Aryal, M. Gupta, and E. Bertino, "A Survey of Agentic AI and Cybersecurity: Challenges, Opportunities and Use-case Prototypes," arXiv:2601.05293, 2026.
[8] S. Barbieri, A. L. R. Ferraz, and L. A. Pereira Junior, "PocketAgents: A Manifest-Driven Library of Autonomous Defense Agents," arXiv:2605.21694, 2026.
[9] C. Guo, G. Pleiss, Y. Sun, and K. Q. Weinberger, "On Calibration of Modern Neural Networks," in Proc. 34th International Conference on Machine Learning (ICML), 2017, pp. 1321-1330. arXiv:1706.04599.
[10] National Institute of Standards and Technology, "Incident Response Recommendations and Considerations for Cybersecurity Risk Management: A CSF 2.0 Community Profile," NIST SP 800-61r3, Apr. 2025. DOI: 10.6028/NIST.SP.800-61r3.
[11] S. Rose, O. Borchert, S. Mitchell, and S. Connelly, "Zero Trust Architecture," NIST SP 800-207, Aug. 2020. DOI: 10.6028/NIST.SP.800-207.
[12] B. E. Strom, A. Applebaum, D. P. Miller, K. C. Nickels, A. G. Pennington, and C. B. Thomas, "MITRE ATT&CK: Design and Philosophy," MITRE Technical Report MP180360, 2018, rev. 2020.
[13] U. Uchibeke, "Before the Tool Call: Deterministic Pre-Action Authorization for Autonomous AI Agents," arXiv:2603.20953, 2026.
[14] G. Zhu and C. Wang, "Intent-Governed Tool Authorization for AI Agents," arXiv:2606.22916, 2026.
[15] B. Wu, W. Zhang, K. Chen, H. Fang, and N. Yu, "Provably Secure Agent Guardrail," arXiv:2605.29251, 2026.
[16] S. V. Kodathala, "aiAuthZ: Off-Host, Identity-Bound Authorization for AI Agents," arXiv:2607.05518, 2026.
[17] National Institute of Standards and Technology, "Artificial Intelligence Risk Management Framework (AI RMF 1.0)," NIST AI 100-1, Jan. 2023. DOI: 10.6028/NIST.AI.100-1.
[18] P. Sharma, "Telemetry Engineering as a Security Architecture Discipline: A Design Framework for Detection-Ready Multi-Cloud Pipelines," manuscript in preparation, 2026.










