AILP — AI ᐃᓕᓐᓂᐊᖅᑎᑦᓯᓂᖅ ᐱᔪᓐᓇᐅᑎᑎᒍᑦ Protocol

AI ᐃᓕᓐᓂᐊᖅᑎᑦᓯᓂᖅ ᐱᔪᓐᓇᐅᑎᑎᒍᑦ Protocol

AI ᐃᓕᓐᓂᐊᖅᑎᑦᓯᓐᓇᖅᐸ ᐅᓇ-ᒥᑦ? ᖃᓄᖅ ᐃᑎᒐᐃᓇᕐᒥᒃ, ᓱᓇᑐᐃᓐᓇᖅ ᐊᑐᖅᑕᐅᓂᕐᒧᑦ, ᓱᓇᒥᒃ ᐱᖁᔭᖅᑕᐅᓪᓗᓂ?

01 · ᑐᑭᓯᐅᒪᓂᖓ

AILP operationalizes the AIRS spectrum for the specific question of learning. It distinguishes acts that binary crawler rules collapse together: being read is not being learned from; being retrieved is not being trained on; fine-tuning is not distillation; verbatim memorization is different from statistical internalization. AILP lets a publisher declare, in a single machine-readable file, per-dimension permissions — access, indexing, inference input, embedding, training, fine-tuning, distillation, memory, output, attribution and compensation — each as allowed, denied, or license-required. An AILP grant answers "may AI learn from this content?" only — it does not answer "may this Agent invoke a tool, write to a database, or publish?" (AARS), nor "who authorized this Agent?" (AADP).

02 · ᑐᕌᖓᔭᖃᕐᓂᖓ

ᐊᐱᖅᓱᖅ "AI ᐃᓕᓐᓂᐊᖅᑎᑦᓯᓐᓇᖅᐸ ᐅᓇ-ᒥᑦ?" ᓇᐅᓇᐃᔭᖅᓗᒍ dimension-level ᓈᒻᒪᒋᔭᐅᓂᕐᒥᒃ crawl bit-ᒥᒃ ᐊᑐᖅᑎᓐᓇᒍ.
ᐃᓕᓐᓂᐊᖅᑎᑦᓯᓂᖅ ᐱᔪᓐᓇᐅᑎᖅ (model-ᒃᑯᓐᓄᑦ ᐃᓚᐅᓪᓗᓂ) ᐊᕕᑕᒍᒃ ᐊᑦᑕᕕᐊ ᐱᔪᓐᓇᐅᑎᒥᑦ (ᑎᑎᖅᖃᑦ fetch-ᑎᒋᓗᒋᑦ).
ᐃᑲᔫᑎᖃᖅᑎᑎᒍᒃ ᐊᑭᓕᖅᑎᑦᓯᓂᖅ model-ᒃᑯᑦ: ᐊᑭᑐᙱᑦᑐᖅ, ᓂᐅᕕᕈᑎᒋᓗᒍ-ᐃᙱᑦᑐᖅ, license ᐱᔭᕆᐊᖃᖅᑐᖅ, ᐊᑭᓕᖁᑎᑦ ᐊᕕᑎᑦᑎᓂᖅ.
ᐊᒻᒪᖅᑐᖅ ᖃᕆᐊᓴᒃᑯᑦ ᐊᑐᖅᑕᐅᓯᓐᓇᖅᑎᑕᐅᒋᐊᖃᖅᑐᖅ — ᓈᒻᒪᒃᑐᖅ ᐃᓱᒪᖃᕐᓂᖅ ᓇᓗᓇᐃᖅᓯᒪᙱᑦᑐᒥᒃ legal-ᒥᒃ ᐱᔾᔪᑎᖃᖏᑦᑐᒥᒃ ᐃᒋᓇᖅᑕᐅᙱᑦᑭᐊᖅ.

03 · Scope

Access / Indexing — Fetch-ᓂᖅ ᐊᒻᒪᓗ Semantic Indexing

inference_input — Context-ᑐᑦ ᐊᑐᖅᑕᐅᓂᖅ Inference ᐅᓪᓗᓕᒫᖓᖓᓂᑦ (RAG)

Embedding — Vectorization ᐊᒻᒪᓗ Retrieval Index-ᒃᑯᑦ

training / fine_tuning / distillation — Model Internalization Tier-ᒃᑯᑦ

verbatim_memory — ᐃᓗᒧᒻᒥ ᓇᕿᑦᑎᓂᖅ ᐊᖏᖅᑕᐅᓂᖃᖅᒪᖔᑕ

Attribution / Compensation — Citation ᐱᓕᕆᐊᖃᕐᓂᖅ ᐊᒻᒪᓗ ᐊᑭᓕᖅᑎᑦᓯᓂᖅ ᓈᓚᐅᑎᒋᔭᐅᔪᖅ

04 · ᖃᕆᐊᓴᒃᑯᑦ ᐊᑐᖅᑕᐅᔪᓐᓇᖅᑐᖅ ᐊᔅᓯᓕᐊᖓ

ᐃᓕᓐᓂᐊᖅᑎᑦᓯᓂᖅ-ᐊᑐᖅᑕᐅᓯᓐᓇᖅᑎᑦᓯᓂᖅ ᓇᓗᓇᐃᖅᓯᓂᖅ — /ai/rights-spectrum.json (AILP profile-ᐊ)

/ai/rights-spectrum.json

{
  "version": "0.1",
  "protocol": "AILP",
  "publisher": "example.org",
  "default": {
    "access": "allowed",
    "indexing": "allowed",
    "inference_input": "allowed",
    "embedding": "allowed",
    "training": "license_required",
    "fine_tuning": "license_required",
    "distillation": "denied",
    "verbatim_memory": "denied",
    "attribution": "required",
    "compensation": "contact"
  },
  "contact": "licensing@example.org"
}

05 · ᑭᓪᓕᖏᑦ

ᐃᓕᓐᓂᐊᖅᑎᑦᓯᓂᖅ-ᐱᔪᓐᓇᐅᑎᑎᒍᑦ ᑐᑭᓯᖁᑎᑦ ᐊᔪᕐᓇᙳᖃᕐᓂᖏᓐᓂᒃ technical-ᑐᑦ ᒥᓱᒃᑭᑦᑎᐊᕐᓂᕐᒧᑦ — AILP ᐃᓱᒪᒋᔮᓂᒃ ᓇᓗᓇᐃᖅᑎᑦᑐᖅ, ᑭᓯᐊᓂᓕ ᓈᓚᙱᓐᓇᒥᖓᓂᒃ ᓱᓕ ᑕᑯᑎᑦᓯᔪᓐᓇᙱᓐᓇᒥᒃ.
Dimension ᐃᓕᑦᑐᕆᕝᕕᐊᖓ draft-ᐅᕗᖅ; ᑭᓪᓕᖏᑦ training-ᒥᑦ, fine-tuning-ᒥᑦ, distillation-ᒥᓗ ᐊᑯᙵᓂ ᓱᓕ ᐅᖃᓪᓚᐅᑎᒋᔭᐅᔪᑦ.
ᒪᓕᒐᕐᔪᐃᑦ-ᑎᒍᑦ ᐃᓕᑕᕆᔭᐅᓂᖅ ᐃᓕᓐᓂᐊᖅᑎᑦᓯᓂᖅ ᐱᔪᓐᓇᐅᑎᓄᑦ ᐊᔾᔨᒌᖏᑦᑐᖅ ᓄᓇᓕᒫᓄᑦ ᐊᒻᒪᓗ ᓵᒻᒪᒃᑎᑕᐅᙱᑦᑐᖅ.
v0.1.1: AILP(Resource,ContentUse)=Allow does not imply AARS(Actor,Action)=Allow, and does not imply AADP(Principal,Actor,Authority)=Valid — a system exposing both content and tools must evaluate them separately.