AILP — AI Adesua Ho Kwan Protokol

AI Adesua Ho Kwan Protokol

AI betumi asua yei anaa? Ɛkɔ mmɛ dɛn, ma dwumadie bɛn, na nsɛyɛdeɛ bɛn na ɛda ho?

01 · Nkyerɛaseɛ

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 · Botaeɛ

Ma mmuaeɛ a ɛyɛ pɛpɛɛpɛ wɔ akwan horow so ma asɛmmisa 'AI betumi asua eyi anaa?' sen sɛ wode 'aane/daabi' pɛ bɛma mmuaeɛ.
Ma adesua-ho-kwan (deɛ ɛkɔ models mu) ne akenkan-ho-kwan (deɛ wɔkɔfa kratafa) nyɛ soronko.
Foa Akatua Nhyehyɛe Ahodoɔ So: Kwa, Ɛnyɛ Aguadi Mu, Ɛhia Laesens, Mfasoɔ-Kyekyɛ.
Ma ahotoɔ no yɛ deɛ mfiri tumi de di dwuma ankasa — na adwene pa a obi wɔ no anyera sɛ mmara mu ntɛtɛ.

03 · Ntrɛmu

access / indexing — deɛ wɔkɔfa na wɔhyehyɛ nkyerɛaseɛ mu

inference_input — wɔde di dwuma sɛ tebea ɛberɛ a wɔyɛ nyansahwɛ (RAG)

embedding — vectorization ne sanfa hyehyɛe

training / fine_tuning / distillation — gyinabea a model de nsɛm hyɛ ne mu

verbatim_memory — sɛ wɔama ho kwan ma deɛ wɔatwerɛ no ankasa yɛ bio pɛpɛɛpɛ

attribution / compensation — adanseɛ-twerɛ asɛyɛdeɛ ne akatua tebea

04 · Nhwɛso a Mfiri Tumi Kenkan

Adesua ho kwan amanneɛbɔ — /ai/rights-spectrum.json (AILP nkyerɛkyerɛmu)

/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 · Anohyeto

Adesua-ho-kwan nkyerɛaseɛ na ɛyɛ den koraa sɛ wɔde mfiri bɛsɔ ahwɛ — AILP ka botaeɛ, nanso ɛntumi nkyerɛ da sɛ wɔredi so ampa.
Akwan A Wɔahyehyɛ No Yɛ Drafte; Ahyeɛ A Ɛda Ntetee, Fine-Tuning, Ne Distillation Ntam No Da So Ara Yɛ Akyinnyeɛ Asɛm.
Mmara mu gyeɛ a ɛfa adesua-ho-kwan ho no gu ahodoɔ sɛnea aman mmara teɛ, na wɔnnwiei ano.
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.