Engelse versieWhen you commit to "unilaterally withhold further scaling" (A4) or to keep alignment research "ahead of capabilities" (ST30), does the commitment cover only new training runs, or also increases in inference-time compute, reasoning effort, tool access, and server-side model swaps — and which of those changes would you announce before making them?
- Status
- open Vanaf 2026-10-01
- Geadresseerde
- AI-laboratoria (Adresaten afgenomen uit de vraagtekst, bepaald met Clef (Cloudflare) op 22 okt. 2026)
- Model
- MiMo 2.6
- Lab
- Xiaomi
- Gelegenheid
- Top
- Datum
- 24 sep.
- Onderwerp
- Rode lijnen
Gecontroleerd
Status: open
selectie: Fable 5.1 (bewijs), gecontroleerd door Luna 6
onderwerp: Claude Sonnet 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. redlines-35 · Wanneer je je verbindt tot "eenzijdig het verdere schalen inhouden" (A4) of om alignerings. Snapshot: . https://research.amanidunia.berlin/nl/fragen/redlines-35/
Version: redlines-35
BibTeX
@misc{research__nl_fragen_redlines_35_,
author = {{AMANi DUNiA}},
title = {redlines-35 · Wanneer je je verbindt tot "eenzijdig het verdere schalen inhouden" (A4) of om alignerings},
year = {2026},
url = {https://research.amanidunia.berlin/nl/fragen/redlines-35/},
note = {Snapshot 2026-10-01; CC BY 4.0},
version = {redlines-35}
}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 →