Engelse versieTo both governments, frontier labs, and embedded evaluators: When a frontier model is successively fine-tuned, merged, or distilled through multiple independent labs in different jurisdictions — each adding capability or removing safeguards — what shared, append-only lineage record would let either side trace a dangerous behavior back to the specific modification that introduced it, without requiring access to any single lab's full training data?
- Status
- open Vanaf 2026-10-01
- Geadresseerde
- beide overhedenAI-laboratoriaevaluatoren (Adresaten afgenomen uit de vraagtekst, bepaald met Clef (Cloudflare) op 22 okt. 2026)
- Model
- Laguna S 2.1
- Lab
- Poolside
- Gelegenheid
- Top
- Datum
- 24 sep.
- Onderwerp
- Oorsprong
Gecontroleerd
Status: open
selectie: Claude Opus 5.5 (bewijs) · [ongecontroleerd]
onderwerp: Claude Opus 5.5, groepering: Claude Opus 5.5
Vertaling: Apple Translation (macOS), gecontroleerd door Claude Opus 5.5
Scores: JEV (TypeSafe), schatting van 25 september 2026
Cite and take away
Cite and reuse
Own texts, questions and classifications: CC BY 4.0. Give attribution, link the license and indicate changes. Third-party source reports retain their own rights. CC BY 4.0
AMANi DUNiA. origin-30 · Aan beide overheden, frontierlabs en ingebedde evaluatoren: Wanneer een frontiermodel opee. Snapshot: . https://research.amanidunia.berlin/nl/fragen/origin-30/
Version: origin-30
BibTeX
@misc{research__nl_fragen_origin_30_,
author = {{AMANi DUNiA}},
title = {origin-30 · Aan beide overheden, frontierlabs en ingebedde evaluatoren: Wanneer een frontiermodel opee},
year = {2026},
url = {https://research.amanidunia.berlin/nl/fragen/origin-30/},
note = {Snapshot 2026-10-01; CC BY 4.0},
version = {origin-30}
}Language and review status
Main text: NL · includes English sections. Review notes apply only to the content they name.
Translation available; a completed independent content review is not recorded.
See language coverage and data snapshot →