Research Paper · Data & AI
Beyond the Model
A Governed Architecture for Consequential AI Systems
Consequential AI requires an architecture that governs evidence, artifacts, evaluation, authority and learning around the model.
Author & Researcher
Steven J. Boyle
Founder of Northline Advisory and author of Northline Research publications and Practitioner Insights on technology leadership, transformation, governed AI systems and responsible implementation.
This page is the authoritative Northline identity record connecting the author, publications, persistent identifiers and public research records used across Northline, ORCID, Zenodo and DataCite.
Research publications
Each publication links to its DOI and preserved Zenodo record.
Research Paper · Data & AI
Artifact-Oriented AI Architecture
Governing State, Authority and Lineage
An implemented architecture for treating durable, versioned and governed artifacts—not transient model responses—as the unit of enterprise AI control.
Research Paper · Data & AI
Evidence Coverage and Traceability
Knowing What Supports a Conclusion
A deterministic evidence architecture for preserving and inspecting the governed relationships that actually support AI-generated conclusions.
Research Paper · Data & AI
Governed Self-Optimization
How an AI System Can Improve Its Work Without Losing Control
A governed optimization loop that lets AI propose and evaluate revisions while deterministic controls retain authority over continuation and promotion.
Research Paper · Data & AI
When Should the AI Stop?
Material Improvement and the Limits of Iterative Optimization
Historical and controlled evidence on material improvement, diminishing returns and the governance of stopping decisions in iterative AI optimization.
Research Paper · Data & AI
Governed Learning
How AI Systems Can Learn Without Turning Experience Into Truth
An evidence-grounded architecture for using decision history as an advisory signal without allowing experience or frequency to become factual authority.
Research Paper · Data & AI
Architecture as a Governing Contract
Engineering Reliable Software When the Implementer Is Probabilistic
An implementation-grounded model for constraining probabilistic software development through deterministic architecture, certification and baseline controls.
Research Note · Data & AI
Preserving Evidentiary Authority During Corrective Regeneration in Multistage LLM Systems
An Applied Engineering Case
An implementation-tested approach to correcting an unauthorized generated claim while preserving the evidence, validated context and acceptance rules governing it.
Research Paper · Data & AI
Neurodiversity-Informed AI Communication Support
Development and Evaluation of an Individualized Calibration Method
A controlled synthetic research programme evaluating whether structured communication calibration can improve individualized AI support recommendations—and where prompt-only execution remains unreliable.
Practitioner Insights
Practitioner Insight · Transformation
You Don't Rise to the Level of Your Strategy — You Fall to the Level of Your Systems
Transformation fails when strategic ambition exceeds the maturity of the systems expected to execute it.
Practitioner Insight · Leadership & Strategy
Technology Readiness Is Becoming Mission Readiness
AI, cybersecurity, modernization and digital service are often treated as separate priorities. In practice, they are testing the same thing: whether the organization can absorb and sustain technology-enabled change.
Practitioner Insight · Governance
AI Governance Isn't a Committee — It's an Operating Model
AI governance cannot be reduced to a committee. Effective governance requires an operating model that connects strategy, architecture, operations, risk, and use-case decisions.
Practitioner Insight · Governance
Architecture Is Not Documentation. It's a Contract.
Architecture becomes valuable when it constrains implementation—preserving decisions, defining boundaries, and preventing locally correct work from violating the system as a whole.
Practitioner Insight · Leadership & Strategy
How to Calculate — and Communicate — the Real Capacity of an IT Service Group
Why coverage, maturity, and staffing depth—not headcount—should drive budget decisions.
Practitioner Insight · Governance
The Executive Responsibility Most Institutions Never Formally Assign
Why system integrity has become a board-level risk—and why modern organizations need clear executive accountability for the digital infrastructure on which they depend.
Practitioner Insight · Governance
Artifact-Oriented AI Architecture: Closing the Enterprise AI Architecture Gap
An architectural pattern for high-consequence enterprise AI systems in which probabilistic reasoning operates within deterministic governance and authority resides in governed artifacts rather than generated responses.
Practitioner Insight · Governance
It's Just RAG. Except That Isn't the Interesting Part.
RAG helps determine what knowledge a model should receive. Governed AI must also determine whether the model should reason at all, whether another intervention is worthwhile, which state remains authoritative, and what the system should learn from the result.
Practitioner Insight · Governance
Governed AI with Gated Learning: Lessons from Building EPS
A governed architecture pattern for introducing learning into consequential AI workflows, illustrated through the development of EPS.
Practitioner Insight · Governance
From Decision History to Governed Intelligence
How governed AI systems can learn from decision history without allowing accumulated experience to silently become authority.
Practitioner Insight · Transformation
Transformation Is a System, Not a Project
What transforming a university credential assessment service taught me about redesigning operations before automating them.
Practitioner Insight · Data & AI
Why Git Matters Even If You Don’t Write Code
A practical account of how Git preserves inspectable work across people, agents and conversations, while governance and discoverability determine what can actually be trusted and found.
Practitioner Insight · Governance
Publishing Independent Research: From Manuscript to Verified Public Record
A practical account of how Northline combined Zenodo, DOIs, ORCID, Git and controlled release checks to make independent research easier to identify, cite and maintain.
Practitioner Insight · Transformation
Alignment at Scale: A Framework for Modern Transformation
Organizations can invest in technology and develop strong departmental strategies without achieving coordinated results. Six interdependent management capabilities help leaders identify what is holding execution back and where intervention matters.