AILP — Protocol y'Olukusa lw'Okuyiga kwa AI

Protocol y'Olukusa lw'Okuyiga kwa AI

AI eyinza okuyiga ku kino? Mu buziba ki, ku bikozesebwa ki, n'obuvunaanyizibwa ki?

01 · Amakulu

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 · Ekigendererwa

Ddamu ekibuuzo "AI eyinza okuyiga ku kino?" mu buneneezi obw'oku mitendera egy'enjawulo, so si kabonero kamu aka funza.
Awula olukusa lw'okuyiga (okuteeka mu mubiri gw'emideeli) okuva ku lukusa lw'okuyingira (okufunza emiko).
Wagira ebifaananyi by'empeera: bwereere, ebitali bya bwenkulungo, ebyetaaga layisensi, n'okugabana amagoba.
Kakasa nti obuggule businza okukolebwa makina — obulungi bw'omutima buleme okusuulibwa ng'obutakakasa mu mateeka.

03 · Ekipimo

access / indexing — okufunza n'okuteeka mu ndagiriro ez'amakulu

inference_input — okukozesa ng'embeera mu kiseera ky'okutegeeza (RAG)

embedding — okufuula ebirimu za vector n'endagiriro z'okuggyayo

training / fine_tuning / distillation — emitendera gy'okuteeka mu mubiri gw'omudeeli

verbatim_memory — obanga okuddamu okufulumya ekituufu kikkirizibwa

attribution / compensation — obuvunaanyizibwa bw'okuwaŋŋula n'ebisaanidde by'okusasula

04 · Ekyokulabirako ekisomeka za makina

Olulanga lw'olukusa lw'okuyiga — /ai/rights-spectrum.json (profayiro ya AILP)

/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 · Ebisaanyizo

Amakulu g'olukusa lw'okuyiga ge gasinga obuzibu okukakasa mu tekiniki — AILP erangirira ekigendererwa, naye tekyayinza kukakasa kugoberera.
Olukalala lw'ebika lwe nteekateeka; enkomo wakati w'okuyigiriza, fine-tuning, ne distillation zikyali mu mpaka.
Okukkirizibwa mu mateeka okw'olukusa lw'okuyiga kwawukana okusinziira ku ggwanga era tekunnagobwaako.
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.