AILP — Protokoli ya Ruhusa ya Kujifunza ya IA

Protokoli ya Ruhusa ya Kujifunza ya IA

IA inaweza kujifunza kutoka hapa? Kwa undani gani, kwa matumizi gani, chini ya wajibu gani?

01 · Ufafanuzi

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 · Kusudi

Jibu swali «IA inaweza kujifunza kutoka hapa?» kwa usahihi wa kila upande, badala ya kitufe kimoja cha kutafuta.
Tenganisha ruhusa ya kujifunza (kuingiza ndani ya modeli) na ruhusa ya ufikiaji (kuchukua kurasa).
Saidia miundo ya malipo: bila malipo, isiyo ya kibiashara, inayohitaji lisansi, kugawana mapato.
Hakikisha uwazi unatekelezeka na masini — ili nia njema isitupwe kama hali ya kutokuwa na uhakika wa kisheria.

03 · Upeo

ufikiaji / uorodheshaji — kuchukua na uorodheshaji wa kimaana

inference_input — matumizi kama muktadha wakati wa hoja (RAG)

embedding — uvekta na orodha za uchukuaji

mafunzo / kurekebisha kwa undani / distillation — madaraja ya kuingiza ndani ya modeli

verbatim_memory — kama uzalishaji sahihi kabisa unaruhusiwa

kutaja chanzo / malipo — wajibu wa unukuzi na masharti ya malipo

04 · Mfano unaosomwa na masini

Tamko la ruhusa ya kujifunza — /ai/rights-spectrum.json (wasifu wa 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 · Vikwazo

Maana za ruhusa ya kujifunza ndizo ngumu zaidi kuhakiki kiufundi — AILP inatamka nia, bado haiwezi kuthibitisha uzingatiaji.
Orodha ya pande bado ni rasimu; mipaka kati ya mafunzo, kurekebisha kwa undani, na distillation bado inajadiliwa.
Utambuzi wa kisheria wa ruhusa za kujifunza unatofautiana kwa mamlaka na bado haujatatuliwa.
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.