HGAA Standard — VERSION 1.2 (May 2026)

Preserving epistemic agency and accountability in AI-mediated work.

Human first. Tools second. Responsibility, always.™

HGAA (Human-Governed AI Authorship)
v1.2 — May 2026 · Expanded for contextual validation, testing criteria, provenance, agency, ethics, and accountability

What This Is (and Why it Matters)

Human-Governed AI Authorship (HGAA) is a standard for using AI without losing human authorship, judgment, contextual validation, or responsibility.

As AI becomes more capable, the risk is not just error. It is the quiet loss of human meaning, voice, context, provenance, and accountability.

HGAA defines the conditions under which AI can be used while an identifiable human source of authorship remains governing, visible, responsible, and accountable for what is created, accepted, and shared.

Read This Plain English Summary

To make this possible in practice, HGAA relies on methods that keep the human source of authorship visible to the system and traceable in the work.

A central condition of HGAA is contextualization, in which AI operates within a clearly authored human context of intent, meaning, purpose, constraint, and judgment. One primary method for achieving this is contextualized augmentation.

Formal Definition

Human-Governed AI Authorship (HGAA) is a standard for AI-augmented authorship in which artificial intelligence may support drafting, language, structure, refinement, or generative extension, while an identifiable human source of authorship remains the governing source of concept, meaning, judgment, approval, and final responsibility for the work.

HGAA establishes the conditions under which AI-augmented work remains legitimately human-authored. AI may extend expression, but it does not replace the human source of meaning, authorship, discernment, or accountability.

Within HGAA, one of the primary methods for preserving authorship is contextualized augmentation: AI operating within a deliberately authored human context, under human judgment and final responsibility.

Traceability ensures that human direction, review, and approval remain visible and, where necessary, demonstrable within the authorship process.

Provenance helps complete this governing process by making human direction, review, and approval visible and traceable. In this sense, provenance supports traceability by connecting technical record-keeping to accountable human authorship.

Scope of the Term

HGAA is intentionally focused on authorship. It applies to AI-augmented writing and other forms of authored expression in which an identifiable human source of authorship must remain responsible for meaning, judgment, approval, and final accountability.

While its principles may inform other areas of AI use, HGAA is primarily concerned with preserving legitimate human authorship in AI-augmented work.

Where human context is thin, AI may produce fluent language without preserving authentic voice, grounded meaning, contextual applicability, provenance, or accountable authorship. HGAA responds by requiring visible human governance, traceable authorship, contextual validation, and authored conditions under which AI remains bounded rather than substitutive.

Plain-Language Version

AI can help carry the expression.
The human must still carry the meaning, judgment, and responsibility.
And that human source of authorship must remain visible and traceable.

Core Conditions of the Term

Human origination or substantive direction
The human source of authorship originates the core idea, or meaningfully directs the conceptual intent, purpose, and direction of the work.
Human discernment
The human evaluates, selects, rejects, revises, or reshapes generated material rather than passively receiving it.
Human contextual validation
The human assesses whether AI-generated output applies to the actual context, has been tested against the governing constraint, and is supported by appropriate provenance before it is accepted or acted upon.
Human final judgment
The human decides what is fitting, what is faithful enough to stand, and what is worthy of being shared.
Human accountability
The human remains answerable for the final work, including its meaning, effects, provenance and public trustworthiness.
Provenance as completion of governance
The work must be traceable to human direction, review, and approval. Provenance is the visible record that governance remained real.

What the Term Rejects

concealed AI ghostwriting presented as wholly human work
the outsourcing of meaning, judgment, or accountability to the model
the assumption that fluency alone is sufficient for authorship
the displacement of human responsibility through technical assistance
the acceptance of AI-generated output without context, constraint-checking, or provenance

Publication-Ready Statement

I define Human-Governed AI Authorship as a standard of AI-augmented authorship in which an identifiable human source of authorship remains the governing source of concept, meaning, judgment, and accountability, while AI serves as a tool of articulation, refinement, and structured support. In this model, AI may extend expression, but it does not assume authorship of the human meaning behind it.

Within HGAA, one of the primary methods for preserving authorship is contextualized augmentation: bounded AI operation within authored human context, under human judgment and final responsibility.

Provenance helps ensure that this governance is not merely claimed, but visible and traceable. Under HGAA v1.2, provenance also supports validation before trust by helping show whether AI-generated output has been grounded in context, tested against the governing constraint, and accepted through accountable human judgment.

Signature Reef Flow Formulation

Human-Governed AI Authorship affirms that AI may extend expression, but human meaning, judgment, and responsibility must remain human-governed, visible, and traceable.

Short Motto

Human first. Tool second. Responsibility always.

Whitepaper Note

This page offers a public-facing presentation of Human-Governed AI Authorship (HGAA) as the governing standard for preserving legitimate human authorship, visible human governance, and accountable judgment in AI-augmented work.

The full framework is developed in the whitepaper Human-Governed AI Authorship (HGAA): A Framework for Preserving Epistemic Agency and Accountability in AI-Augmented Work, which explains in greater depth the conditions under which human judgment, accountable authorship, and legitimate human governance remain visible and intact in AI-augmented work.

Request your copy here.

Positioning Note

HGAA is a public-facing authorship standard for preserving legitimate human authorship in AI-augmented writing. It is designed to work alongside broader AI governance resources, including NIST’s voluntary AI risk framework, by translating concerns such as traceability, transparency, and human oversight into the specific domain of authorship.

HGAA also operates alongside current U.S. Copyright Office guidance on human authorship by helping writers preserve and demonstrate meaningful human judgment, accountable authorship, and final responsibility in AI-augmented work.

Permission Statement

Permission is granted to share this document for non-commercial, educational purposes, provided it is circulated with clear attribution to Mr. Ed Woods / Reef Flow Publishing and is not presented as the work of others.

ReefFlowPublishing.com | ARCframework.ai | HumanGovernedAI.com

Human first. Tool second. Responsibility always.

Reasons to use the arc framework

accurate and reliable Results

The ARC Framework™ is a two-tier, curation continuum process that transforms AI from a content generator into a contextual collaborator.

Tier 1 ensures each output is accurate and reliable—by validating the AI’s role, sources, and citations.
Tier 2 refines those outputs into strategic, contextual decisions—by analyzing risks, reframing assumptions, and committing to action.

Together, these tiers anchor AI insights in human relevance—so you don’t just get answers, you get results you can trust.

All assets published under ARCframework.ai, including the ARC Framework™, Reef Flow™ narratives, ARCemedes™, Qwilla AI™ and the ARCframework-OpenAcademicEdition GitHub repository, are protected intellectual property. The framework content is licensed under CC BY-ND 4.0 for educational and nonprofit use only. Reef Flow’s characters, metaphors, and narrative structure are trademarked and may not be reproduced, adapted, or distributed without express permission. Commercial use of any ARC-aligned tools or materials requires a separate license. Please see our Licensing page for full terms.

Mr Ed Woods - ARCemedes Master Steward
(727)-470-6420
Clearwater, Florida USA
www.MrEdWoods.com
ARCF LInkedIn Business