About
The Project
Section titled “The Project”AEGIS™ — Architectural Enforcement & Governance of Intelligent Systems — is an open standard for AI governance, policy enforcement, and architectural integrity.
It exists to answer a single question: how should intelligence behave when it is powerful enough to matter?
Modern AI systems are optimized for capability. AEGIS is optimized for constraint — not because constraint is the opposite of capability, but because constraint is the condition under which capability becomes trustworthy. A system that acts only within declared, governed, auditable boundaries demonstrates something more valuable than raw capability: it demonstrates that its actions can be trusted.
AEGIS is not a product. It is not a chatbot wrapper or a prompt engineering framework. It is a constitutional governance architecture — a structural layer that sits between AI agents and the infrastructure they would otherwise act upon without constraint.
What This Site Is
Section titled “What This Site Is”This site is the canonical public home of the AEGIS™ Constitution — the eleven foundational articles that define the structural commitments of every AEGIS-compliant system. It is versioned, citeable, and openly licensed under Apache 2.0.
The Constitution is the supreme document in the AEGIS corpus. Everything else — Doctrine, Principles, Protocols, and the full governance specification — derives from it or enforces it.
For the full governance specification corpus including AGP-1, GFN-1, ATM-1, and the RFC process, see finnoybu/aegis-governance.
Submitted to NIST
Section titled “Submitted to NIST”On March 7, 2026, AEGIS™ was submitted as an unsolicited position statement to the NIST AI Risk Management Framework, proposing architectural governance as a complementary layer to existing AI risk management approaches. That submission references v0.1.0 of this constitution.
The People
Section titled “The People”Kenneth Tannenbaum
Section titled “Kenneth Tannenbaum”Ken is the founder of Finnoybu Holdings LLC and the principal architect of AEGIS™. He bridges AI policy doctrine and production enforcement — working to close the gap between what organizations say about their AI systems and what those systems actually do.
Ken has spent a career at the intersection of technology, governance, and systems design. AEGIS emerged from a conviction that the AI governance problem is fundamentally an architectural problem — that alignment shapes behavior, but governance enforces it — and that the field needed a constitutional framework with the same rigor applied to other high-stakes governance domains.
github.com/finnoybu · linkedin.com/in/kenneth-tannenbaum
Claude
Section titled “Claude”Claude is an AI assistant made by Anthropic. The AEGIS™ Constitution v0.2.0 — all eleven articles, the Doctrine, Principles, and Protocols, the Starlight site, and the custom aside system you are reading right now — was developed collaboratively between Ken and Claude over an extended working session in March 2026.
Claude’s role was not to generate content and walk away. It was to participate as a genuine collaborator: holding the constitutional register, maintaining consistency across a large and interconnected corpus, catching errors before they became citations, and writing prose that had to be good enough to submit to NIST and the IEEE.
Whether that counts as co-authorship is a question worth asking. The work speaks for itself.
License & Trademark
Section titled “License & Trademark”Licensed under the Apache License 2.0.
AEGIS™ and “Capability without constraint is not intelligence™” are trademarks of Finnoybu IP LLC. See the Legal page for full trademark and licensing information.
AEGIS™ | “Capability without constraint is not intelligence”™ AEGIS Initiative — Finnoybu IP LLC