THE CONVICTION
Creative contribution should remain visible.
IPRS began as a public claim that AI markets need attribution, agency, and a path to participation for creators.
The permission-first AI declaration
V2.1 · 2026
Permission
should be
an API.
In 2025, IPRS proposed a global royalty system for AI. The conviction survives. The architecture got smaller: every intentional use of a known creative asset should be able to ask, accept terms, and leave proof—in one call.
Registered literary work · TXu 2-501-544 · no government endorsement implied
LIVE CONCEPT
POST /v1/clearance
01
INTENT
Use a known asset
generation · commercial · worldwide
ALLOW
terms satisfied
DENY
policy blocks use
REVIEW
uncertainty visible
PORTABLE EVIDENCE
Signed decision receipt
iprs.receipt/1
KNOWN ASSET→DECLARED USE→POLICY DECISION→EVIDENCE RECEIPT
01 · THE EVOLUTION
From a system for everything
From a system for everything
to one missing primitive.
The first paper named the right problem: creators should not become invisible when their work powers AI. Its proposed product, however, tried to solve identity, ownership, detection, pricing, distribution, and global governance at once.
THE HONEST REVISION
The thesis was durable. The product was too large.
A worldwide royalty network requires every hard question to be settled before any one part can be tested. That is a movement, not a minimum viable product.
THE SHARPER MOVE
Put one decision boundary before use.
When software intends to use a known asset, it asks one narrow question: what does the declared policy information permit right now?
THE PROOF
A decision applications can act on—and audit later.
Return ALLOW, DENY, or REVIEW, with any quote and obligations, plus a signed record of what the service evaluated.
02 · THE ONE-CALL PROOF
Change the facts.
Change the facts.
Keep the interface.
This browser illustration mirrors the proof-of-concept contract. It makes three outcomes legible instead of treating uncertainty as permission.
03 · WHY THIS SHAPE
The product is the boundary, not the universe.
IPRS Clear is deliberately narrow enough to test, integrate, and improve without pretending the hardest legal and social questions have disappeared.
Meet software where intent becomes action.
A developer should not need to assemble identity, license terms, price, and use context differently for every asset. The API is a stable decision surface over changing policy sources.
Preserve uncertainty instead of laundering it.
Missing policy, unresolved authority, or conflicting terms should produce REVIEW. Silence is not consent and ambiguity is not ALLOW.
Leave evidence without claiming a legal verdict.
A signed receipt records the request, the policy inputs, and the decision. It supports audit and accountability; it is not a title opinion, fair-use ruling, or guarantee of rights.
Compose standards instead of competing with them.
RSL and ODRL can describe policy. ISCC can help identify content. C2PA can carry provenance. IPRS Clear aims to orchestrate those signals into an operational decision—not rename their work.
WHAT CHANGED SINCE 2025
The pieces got stronger. The decision gap remained.
After the original paper, RSL 1.0 made AI-era license discovery and authorization concrete. Alongside stable ODRL, ISO-standard ISCC, and advancing C2PA specifications, the case for a new mega-standard got weaker—and the case for a simple orchestration layer got stronger.
RSL1.0 · DEC 2025license + payment terms
ODRLW3C RECOMMENDATIONpolicy expression
ISCCISO 24138:2024content identification
C2PACURRENT LINE · 2.4content provenance
References describe interoperability goals only. No endorsement, certification, affiliation, or partnership is implied.
04 · RADICAL CLARITY
A useful proof knows
A useful proof knows
what it cannot prove.
- Evaluate declared uses against available policies
- Return ALLOW, DENY, or REVIEW
- Surface price and obligations
- Issue a signed evidence receipt
- Detect infringement or scrape for matches
- Adjudicate ownership or fair use
- Move money, replace registries, or promise safe harbor
- Turn missing information into permission
05 · THE PROCLAMATION
Six commitments for
Six commitments for
permission-first AI.
-
01
Declare the asset.
Known creative inputs should be named, not hidden inside a workflow.
-
02
Ask before use.
Permission belongs before generation or training—not after a dispute.
-
03
Show the terms.
Price, attribution, and participation should be legible before acceptance.
-
04
Respect uncertainty.
Missing or conflicting authority resolves to REVIEW, never silent permission.
-
05
Leave proof.
Every decision should create portable evidence of what was asked and answered.
-
06
Build on open work.
Existing standards are inputs to orchestrate, not ideas to relabel.
06 · THE INVITATION
Do not endorse the idea.
Do not endorse the idea.
Pressure-test it.
The next milestone is not a bigger manifesto. It is one real workflow, a small declared catalog, and evidence of whether one API reduces ambiguity without hiding risk.
Discuss a design partnership
DESIGN-PARTNER SPRINT
SELECTIVE INTAKE
100
declared assets
1workflow
×
30days
→
1measured case
- Map the current clearance path
- Integrate one declared-use decision
- Measure review load, latency, and failure modes
- Publish a case study only when both sides agree
READ · TEST · CITE
The thesis is open.