CORTEXA
← Browse
openalexZenodo (CERN European Organization for Nuclear Research)2026-07-23Cited by 0

A Cognitive Defense Framework for Detecting and Containing Autonomous Cyber Incidents Caused by Next-Generation Agentic Artificial Intelligence

Dmytro Prokopovych-Tkachenko

Abstract: The rapid maturation of agentic artificial intelligence (AI) is producing a class of cyber threats in which an autonomous system can plan action sequences, adapt its strategy to intermediate results, and interact with networked resources without direct human oversight. Conventional signature-, anomaly-, and event-based defenses respond to isolated indicators and are poorly suited to machine-initiated behavior that unfolds as many coordinated, low-visibility operations. This work proposes a conceptual cognitive-defense framework aimed at detecting, forecasting, and containing autonomous cyber incidents attributable to agentic AI. Rather than matching individual signatures, the framework reconstructs an agent's inferred intent, goal structure, and causal action chain, and compares the declared task objective against observed behavior. The architecture combines continuous runtime monitoring, semantic command analysis, a causal action graph, an ensemble of independent observer models, and a graded response mechanism spanning privilege reduction, process suspension, token revocation, and network isolation. A Cognitive Autonomous Threat Index (CATI) aggregates plan complexity, action interdependence, stealth, adaptivity, privilege-escalation attempts, and persistence potential into a single risk score. The framework is positioned against recent digital-twin, edge-integrity, and cyber-resilience studies, and an illustrative scoring walkthrough demonstrates its operation. Empirical validation on an instrumented testbed is identified as the principal direction for future work. Keywords: agentic AI; cybersecurity; autonomous threat; intent reconstruction; cognitive defense; cyber–physical systems; digital twin; cyber resilience; runtime monitoring; ensemble detection

View free PDFSource page

Related papers

openalexZenodo (CERN European Organization for Nuclear Research)

Automatic classification of cyber incidents using privacy-preserving artificial intelligence

Loya Caroldene Haughton, Eduardo Fidalgo, David Lewis

As cyber incidents increase in complexity, diversity and frequency, cybersecurity practitioners find it more challenging to extract meaningful threat intelligence and insights from cyber incident reports. This problem is worsened by the limited number of these reports; as cyber i…

Also available via: European Organization for Nuclear Research

View free PDFSource page
openalexZenodo (CERN European Organization for Nuclear Research)2026-07-23

EL-RAKHAWI SOVEREIGN JURISPRUDENCE OF ARTIFICIAL INTELLIGENCE LIABILITY A Comprehensive Architecture for Criminal and Civil Responsibility in the Age of Autonomous Systems, Embodied Robotics, and Artificial General Intellig

m el-rakhawi

EL-RAKHAWI SOVEREIGN JURISPRUDENCE OF ARTIFICIAL INTELLIGENCE LIABILITY A Comprehensive Architecture for Criminal and Civil Responsibility in the Age of Autonomous Systems, Embodied Robotics, and Artificial General Intellig

View free PDFSource page
openalexZenodo (CERN European Organization for Nuclear Research)2026-07-26

Narsi Regression and Narsi Intelligence: A Unified Theoretical Framework for Dynamic Representation Evolution, Recursive Cognitive Adaptation, and Self-Evolving Artificial Intelligence

A Chaudhary

Narsi Regression is a theoretical framework that extends conventional machine learning by treating representation evolution as an explicit optimisation problem rather than an implicit consequence of parameter optimisation. The framework models a learner as a dynamic state consist…

View free PDFSource page
openalexZenodo (CERN European Organization for Nuclear Research)2026-07-24

AUTONOMOUS ROBOTICS THROUGH DETERMINISTIC (NON-ITERATIVE) ARTIFICIAL INTELLIGENCE: AUTONOMY THROUGH STRUCTURE. INTELLIGENCE THROUGH DETERMINISTIC EQUILIBRIUM.

H.M. Cekirge

Artificial intelligence has achieved remarkable success through optimization-based learning, probabilistic reasoning, deep neural networks, and increasingly complex computational architectures. Despite these advances, many autonomous robotic systems continue to rely on iterative…

View free PDFSource page
openalexZenodo (CERN European Organization for Nuclear Research)2026-07-25

# Artificial Intelligence in Metallurgical Engineering: A Comprehensive Review of Applications, Challenges, and Future Direction

Sudhakar Geruganti

## ALTERNATIVE TITLES ### Alternative Title 1 (Comprehensive)**"AI-Driven Transformation in Metallurgical Engineering: From Microstructure Analysis to Smart Manufacturing and Sustainable Production"** ### Alternative Title 2 (Process-Focused)**"Machine Learning and Deep Learning…

View free PDFSource page
openalexZenodo (CERN European Organization for Nuclear Research)2026-07-24

DETERMINISTIC SUSTAINABLE COST-EFFICIENT ARTIFICIAL INTELLIGENCE: A Complementary Computational Framework

H.M. Cekirge

Deterministic Sustainable Cost-Efficient Artificial Intelligence presents a complementary deterministic (non-iterative) perspective on artificial intelligence. Rather than treating learning exclusively as an iterative optimization process, this book explores conditions under whic…

View free PDFSource page