Research papers, formal publications, and written thought leadership. Use the keyword filters or the knowledge graph to narrow the list.
AiSuNe Foundation — registered research and advisory organization
Publications
Concept Graph
Click a node to filter by concept. Click again or the background to clear.
Node size reflects the number of papers covering that concept.
Publications
2026
-
Governance Twin Framework: Sentinel+Council Architecture for Agentic AI Systems in Digital Twin-Enabled Predictive Maintenance
Presents the Governance Twin framework with Sentinel+Council architecture applied to agentic AI systems operating within digital twin-enabled predictive maintenance environments. Accepted for publication and presentation, co-authored with N. P. Greis and R. Doten.
-
AI Governance Already Has a Track Record. We Just Haven't Read It.
Brief arguing that AI governance can draw directly on constitutional history: the same five Governance Twin functions (sentinels, council, historian, evidence bundles, EMV) appear across centuries of polity-scale governance and can be lifted into runtime governance for agentic AI. Concept DOI: 10.5281/zenodo.20125757.
-
Understanding AI Governance Through Apprenticeship: The Zen of Governing Through Guidance Rather Than Control
Reframes AI governance as apprenticeship: agentic systems are governed through ongoing guidance and example rather than static rules. Establishes the AI Governance Twin as the runtime mechanism that makes guidance-based governance practical. Also available on SSRN (6597019).
-
Intrinsic Reliability and Robustness for Hyper-Complex Agentic AI Systems: Solution Outline — Architecture and Strategy
Architecture-and-strategy outline for an intrinsically reliable agentic system: sentinels, council, historian, evidence bundles, and EMV combined into the AI Governance Twin blueprint. Also available on SSRN (6631438).
-
Short-Term Gain, Long-Term Fragility: AI Labor Substitution and the Erosion of Sustainable Capability
Examines how aggressive AI labor substitution generates short-term productivity gains while eroding the human capability stock organizations need long-term. Frames capability decay as a governance-relevant form of drift. Also available on SSRN (10.2139/ssrn.6577818).
-
Intrinsic Reliability and Robustness for Hyper-Complex Agentic AI Systems: Governance Twin with Sentinels and Council
Extends the intrinsic reliability work by centering the Governance Twin together with Sentinels and Council as the runtime governance architecture for hyper-complex agentic systems.
-
Delegation Without Living Governance
Argues that governance frameworks built on predefined rules, post-hoc accountability, and static compliance break down once judgment is delegated to agentic AI operating at runtime and machine speed. Introduces runtime governance and the Governance Twin as continuous oversight mechanisms that preserve human relevance during execution.
-
Why AI Threats Make Post-Quantum Cryptography Urgent Today
Frames post-quantum cryptography as a coupled governance and security requirement rather than a separate infrastructure upgrade. AI acceleration increases cryptographic risk across model systems, agent communication, and governance infrastructure, making PQC an immediate design concern.
2025
-
The Governance Twin for Intrinsically Robust and Reliable Agentic AI
Presents the Governance Twin as a parallel runtime architecture for agentic systems whose decisions emerge dynamically. The model separates governance from execution so behavioral drift can be monitored in real time, interventions can happen at system speed, and meaningful human oversight remains possible.
-
Intrinsic Reliability and Robustness for Hyper-Complex Agentic AI Systems: From Pattern Detection to Collective Intelligence
Addresses the reliability problem created by high autonomy and system complexity. Introduces a Sentinel-Council architecture, together with Minimum Viable Ethics, to monitor evolving behavior, define non-negotiable boundaries, and sustain robust performance under changing conditions.
-
Ethics, Morals, and Values: A Framework for Real-World Outcomes
Defines ethics as constraints, morals as reasoning, and values as prioritization in order to make ethical language operational inside real systems. The framework aims to translate philosophy into design choices, decision logic, and enforceable governance mechanisms.
-
AI as Alien Intelligence: A Relational Ethics Framework for Human-AI Co-Evolution
Argues that static ethical frameworks cannot adequately govern advanced AI systems. Proposes a relational ethics approach treating AI as fundamentally alien intelligence, introducing Minimum Viable Ethics as flexible foundational principles for human-AI co-evolution.
-
From Principles to Relationships: Redesigning Ethics for AI's Alien Cognition
Reframes AI ethics away from static, human-centric rules toward relational ethics. Alien cognition cannot be governed solely through predefined principles; ethics must instead emerge through interaction, adaptation, and ongoing management of the human-AI relationship.
-
Expert Commentary: Iberian Peninsula Power Outage and Grid Resilience
Expert commentary on the April 2025 Iberian blackout, analysing how poor grid integration — not interconnection — created an energy island vulnerable to cascading failure, with implications for critical infrastructure governance.
-
DeepSeek and the Future of GenAI
Examines DeepSeek as a signal of where generative AI may be headed, with emphasis on strategic implications for capability development, competition, and governance.
2024
-
Neurodiversity and Language Skills: The New Superpowers in Generative AI and Prompt Engineering
Explores neurodiversity and language capability as strategic strengths in generative AI and prompt engineering, positioning human cognitive variation as an advantage in shaping effective human-AI interaction.
No papers match the current filter.