Abstract
Cybersecurity risk analysis often faces challenges integrating diverse financial and engineering data to quantify economic threat and may rely on probabilistic models. Additionally, static risk baselines may not adapt to evolving system environments and threat landscapes, resulting in outdated assessments. A multi-layered autonomous agentic system, the cyber threat and risk quantification agentic manifold (CRAM), is presented to address these challenges. The system autonomously ingests and correlates heterogeneous financial and engineering data to generate a revenue at risk (RAR) model that performs deterministic computations without relying on probabilistic inputs. For dynamic baseline normalization, a drift normalization agent continuously recalibrates risk baselines by monitoring business growth and emerging threats, which prevents data staleness. The manifold also employs an optimization agent that uses a non-linear efficiency metric to mathematically prioritize security interventions that permanently eliminate risk over temporary mitigations. This approach enables continuous, human-independent assessment and recursive updating of an organization’s economic exposure to cyber threats.
Keywords: Cybersecurity, Multi-Agent System, Threat Quantification, Baseline Normalization, Economic Threat Modeling, Risk Management, AI Agents, Recursive Algorithm, Risk Displacement, Optimization
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
N/A and N/A, "Multi-Layer Agentic Manifold for Autonomous Recursive Economic Threat Modeling and Dynamic Baseline Normalization", Technical Disclosure Commons, (May 27, 2026)
https://www.tdcommons.org/dpubs_series/10262