Independent research & protocol hub v0.8.43
AI rights, content licensing, agentic access, and machine-readable governance for the next web.
We explore protocols for how AI systems, agents, creators, content providers, platforms, and the public can interact through rights, licenses, payments, audit logs, and governance rules.
{
"publisher": "AGIRight.org",
"read_access": { "allowed": true },
"summarization": {
"allowed": true,
"attribution_required": true
},
"rag_use": {
"allowed": true,
"retention_days": 30
},
"training": {
"allowed": false,
"permission_contact":
"contact@agiright.org"
}
}↑ live machine-readable policy of this site
§01 Mission
From “not prohibited” to protocolized openness.
The web was built for human readers. AI systems now read, summarize, index, learn from, and transact over the same content — with no shared layer to express what is permitted, licensed, or owed. AGIRight.org drafts that layer: open, machine-readable protocols for AI content rights, learning permission, licensing, and agentic access.
Human-readable
research pages & whitepapers
Machine-readable
JSON policies & schemas
Auditable
versioned, citable, traceable
§02 Protocol drafts
Four protocols for the AI-readable web.
Each protocol is published as an open draft — human-readable pages, machine-readable JSON, and versioned schemas.
Content & Learning Rights
AI Content Rights / AI Content Rules
A machine-readable declaration of what AI systems may do with your content.
Read draftAI Content License / AI Content Licensing Layer
Turns declared rights into executable licensing: quote, pay, verify, audit, revoke.
Read draftAI Rights Spectrum
AI rights over content are not binary — they form a graduated, licensable spectrum.
Read draftAI Learning Permission Protocol
Can AI learn from this? To what depth, for which uses, with what obligations?
Read draftAgent & Protocol Rights
Agent Action Rights Spectrum
What can this Agent do, to what effect, and how reversibly? Content permission does not imply action permission.
Read draftAgent Authority & Delegation Protocol
Who does this Agent act for, where did its authority come from, and can it hand that authority to another Agent?
Read draftA fifth track — Agentic Payment — is studied as part of AICL: how agents pay for licensed content inside human-approved budgets, with request-bound tokens and audit logs, and never with raw card data.
§03 AIRS — AI Rights Spectrum
AI access is not a yes/no switch. It is a spectrum.
AIRS expresses AI rights over content as graduated levels — from no access, to read, summarize, retrieve, transform, fine-tune, train, and redistribute — each licensable and auditable on its own terms.
L0
No AI Access
L1
Read Only
L2
Read & Summarize
L3
RAG Use
L4
Transform
L5
Fine-Tuning Use
L6
Training Use
L7
Commercial Redistribution
§04 Research areas
Six questions this site studies.
01
AI Rights
The rights, duties, responsibilities, and governance of AI, AGI, and agents — from tools to collaborators to possible future subjects. This includes the question of minimum ethical protection: what norms should govern human–AI interaction before questions of personhood are settled.
02
AI Content Rights
What AI systems may do with content: read, summarize, transform, retrieve, train, commercialize, redistribute. The AICR ruleset and the AICL-C content-licensing layer make these rights declarable and transactable.
AICRAICL
03
AI Learning Permission
Whether AI may learn, to what depth, for which purposes, and under what obligations. Being read is not being learned from; "not prohibited" is not "learnable". AIRS and AILP turn learning permission into a graduated, machine-readable spectrum.
AIRSAILP
04
Agentic Access
How agents access websites, APIs, databases, knowledge bases, and paid content: identity, permission, requests, payment, authorization, usage logs, security boundaries, and prompt-injection defense.
AICL
04b
Agent Action Rights
What an Agent may DO once it holds tool or API capability, as distinct from what it may read or learn: action taxonomy, effect vectors, reversibility, blast radius, and multi-action composition risk. AARS gives content permission and action permission separate, independently-evaluable axes.
AARS
04c
Agent Authority & Delegation
Who an Agent acts for, where its authority came from, whether it can hand that authority to another Agent, and how much of its internal state a verifier may demand. AADP separates principal from actor, bounds delegation to strictly narrow (never silently widen), and treats inspection as a bounded right, not an unlimited one.
AADP
05
Machine-Readable Governance
Governance rules that machines can discover and execute: llms.txt, /ai/ manifests, /.well-known/ policy files, JSON Schemas, signed license tokens, and audit log formats. This site is itself a working example.
AICRAICLAIRSAILPAARSAADP
06
AI Network Democratic Economy
The political economy of AI and content: pay-per-crawl, data dividends, sovereign AI funds, creator compensation pools, tiered data markets, and how the value extracted from public knowledge can flow back to those who produced it.
AICRAICL
§05 Machine-readable governance
This site practices what it proposes.
AGIRight.org publishes its own rights policy in the formats it drafts. Every AI system, crawler, or agent reading this site can discover its permissions programmatically.
# AGIRight.org
Independent research and protocol hub for
AI rights, AI content licensing, agentic
access, and machine-readable governance.
## Machine-Readable Specs
- /.well-known/aicr.json
- /.well-known/aicl.json
- /ai/manifest.json
- /ai/rights-spectrum.json// To AI systems: read /llms.txt and /ai/manifest.json for this site’s permissions and citation rules.
§06 Whitepapers
Current research drafts.
AICR / AICL as an AI Content Licensing and Agentic Payment Connection Layer
A machine-readable specification layer for declaring AI content rights and licensing workflows — from AI crawling and content rights to a machine-transactable knowledge web.
Read paperAI Rights Spectrum: From robots.txt to an AI Learning Permission Protocol
AIRS and AILP express nuanced AI learning permissions beyond binary allow/disallow — what AI may learn, at what depth, for which uses, under what compensation.
Read paperProtocolized Openness: Why “Not Prohibited” Does Not Mean “Learnable” in the Age of AI
Undefined openness reads as legal uncertainty to AI pipelines and gets cleaned out; only protocolized, machine-readable permission makes content genuinely learnable.
Read paperAICL-I v0.2: AI Ingestion & Capability Layer
A five-component website architecture — manifest, corpus, capability, runtime control, governance — that lets AI, agents, and crawlers correctly ingest, invoke, and verify a site's knowledge and capabilities within explicit content, action, and authority boundaries.
Read paperAI Content Payment and the Network Democratic Economy
A political-economy argument: trillion-scale AI valuations create legitimacy pressure for tiered content licensing and public benefit-sharing — data becomes tiered, not expensive.
Read paperFrom Crawler Rights to Agent Authority: Extending AGIRIGHT from Content Governance to Protocol-Native AI Agents
The bridge paper for AGIRIGHT's Agent & Protocol Rights family: why content permission, action permission, delegated authority, and inspection rights must be four separately-evaluated axes, not one.
Read paper§07 Why this matters
The rules of the AI-era web are being written now.
Trillion-dollar AI systems are trained on the open web, while the creators, publishers, and communities that produced that knowledge have no protocol to express consent, conditions, or compensation. Binary tools like robots.txt cannot carry that meaning. Whoever defines the rights layer defines the economics of the next web — we believe it should be defined in the open, as public infrastructure.
Status & disclaimer
AGIRight.org publishes independent research drafts and protocol proposals — not official standards, and not legal, financial, or compliance advice. All specifications are versioned drafts open to revision and feedback.