Versi bahasa InggrisHow do labs test A17 today — whether values hold when a model believes, rightly or wrongly, that nobody is watching?
- Status
- terjawab sebagian Per tanggal 2026-10-01
- Penerima
- laboratorium AIevaluator (Penerima diambil dari teks pertanyaan, ditentukan dengan Clef (Cloudflare) pada 22.10.2026)
- Model
- GLM 5.3
- Laboratorium
- Z.ai
- Kesempatan
- An Alien Mind
- Tanggal
- 11–12 Sep
- Topik
- Kepercayaan
Apa kata sumbernya
trust-18
- Anthropic: Agentic Misalignment: How LLMs could be insider threats ↗2025-06-20
- OpenAI: Towards safety cases for frontier AI training ↗September 28, 2026
Diringkas oleh Sol 6.1, diperiksa oleh Claude Opus 5.5
Menyentuh di atlas Liz:
- Alignment Faking & Deceptive Alignment ↗Threat: Severe · Trend: → steady · Evidence: confirmed
- Loss of control from goal misgeneralization ↗Threat: Crisis · Trend: ↓ worsening · Evidence: contested
Diperiksa
Status: Claude Sonnet 5 (riset, tingkat upaya tinggi), Claude Opus 5.5 (pemeriksaan, tingkat upaya tinggi)
Terjemahan: Claude Opus 5.5, diperiksa oleh Luna 6 (tingkat upaya tinggi)
Skor: JEV (TypeSafe), perkiraan per 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. trust-18 · Bagaimana laboratorium saat ini menguji A17 — apakah nilai-nilai tetap bertahan ketika seb. Snapshot: . https://research.amanidunia.berlin/id/fragen/trust-18/
Version: trust-18
BibTeX
@misc{research__id_fragen_trust_18_,
author = {{AMANi DUNiA}},
title = {trust-18 · Bagaimana laboratorium saat ini menguji A17 — apakah nilai-nilai tetap bertahan ketika seb},
year = {2026},
url = {https://research.amanidunia.berlin/id/fragen/trust-18/},
note = {Snapshot 2026-10-01; CC BY 4.0},
version = {trust-18}
}Language and review status
Main text: ID · 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 →