This work proposes ARIA as a finance-specific reference architecture and falsifiable research agenda for agent-population governance and organizes six capabilities across normative-accountability, execution-control, and assurance-learning planes: policy specification, population-level observed-versus-expected behavior monitoring (M2), bounded authority, runtime containment, adaptive policy change, and preserved human oversight competence.
Abstract
Financial institutions are beginning to deploy agentic workflows in credit, fraud, collections, compliance, and operational control. Governance remains largely component-centric: each model or agent is specified, tested, authorized, and monitored locally. That is insufficient when institutional risk arises from the joint behavior of many locally acceptable components. We call this gap constitutional non-compositionality: local compliance checks need not compose into acceptable collective outcomes such as bounded disparate impact, market integrity, or traceable accountability. We propose ARIA as a finance-specific reference architecture and falsifiable research agenda for agent-population governance. It organizes six capabilities across normative-accountability, execution-control, and assurance-learning planes: policy specification, population-level observed-versus-expected behavior monitoring (M2), bounded authority, runtime containment, adaptive policy change, and preserved human oversight competence. Two simulations illustrate shared-signal thin-file exclusion under local controls and earlier warning from observed-versus-expected distributional monitoring in a constructed drift regime. The contribution maps these controls to fair-lending, EU AI Act, model-risk, and conduct-supervision evidence needs, and closes with a validation agenda rather than a production-effectiveness claim.
LATTICE (Layered Agentic Triad Topology for Intelligent Coordinated Execution), a governance-first architecture that reframes the authorization question from “do the authors trust this AI?” to “do they trust this architecture?”
Elias Calboreanu· Frontiers in Artificial Inte...· 2 citations
Enterprise digital twins are evolving from passive monitoring replicas into agentic, decision-capable cyber-physical intelligence layers that can observe, reason, plan, and act across complex operational ecosystems. However, as autonomy increases, static governance policies become insufficient to manage risk, drift, co...
Rakesh Kumar Agrawal· International Journal of Glo...· 0 citations
This article compares US, Canadian, and EU regulatory approaches and shows how divergence creates compliance, competitive, and cross-border risks, and proposes four controls: autonomy-proportional governance, security by design, agent stress testing, and fiduciary controls.
Vincent Grégoire· Journal of Portfolio Managem...· 0 citations
This work develops a multilevel governance theory for agentic AI and test its mechanisms in three studies over nine model versions, from a three-billion-parameter local model to a commercial frontier system.
SAGE-Fin is presented, a finance-specific authority-handoff contract that makes the proposed effect, not merely its text, the object of runtime control, and its results establish executable conformance, not independent safety accuracy.
Rui Tang, Qiang Liu, Yi-Chi Zhang et al.· 1 citation
Production AI systems typically apply governance—compliance checking, auditing, and explanation—as an external control layer that observes outputs after they are produced. In this paper, embedded governance is proposed: a governance gate that forms part of the system architecture itself, positioned between inference an...
Waleed A. Al Shehri· International Journal of Adv...· 0 citations