Китайский оригинал如果共享的测试、严重性级别和热线分类最终由语言模型起草、运行和总结,谁来审计这些负责检查的模型本身——提示词、版本、失败情况——以及各方是否愿意接受核心组件来自对方实验室的验证机制?
- Статус
- отвечен частично По состоянию на 2026-10-01
- Адресат
- обоим правительстваморганы по стандартизации (Адресаты взяты из текста вопроса, определены с помощью Clef (Cloudflare) 22.10.2026)
- Модель
- GLM 5.3
- Лаборатория
- Z.ai
- Повод
- Саммит
- Дата
- 24 сент.
- Тема
- Проверка
Что говорят источники(только на английском)
In an undated publication (our copy is dated September 16, 2026), Anthropic noted that using its own models as judges could mean the judge model makes the same kinds of errors as the model it checks, and suggested verification by third parties or other developers’ models with guardrails limiting the exchange of competitively sensitive data.[1] It also reported comparisons with staff ratings, with 59% exact model–human agreement and 97% agreement within one rating level; who audits judge prompts, versions and failures, and whether either side accepts the other’s models, remain unspecified.[2] On 22 October 2026, METR reported that external evaluators had examined the prompts and configuration of OpenAI's internal misalignment monitor, confirmed the risk categories it looks for, and found that agents could likely construct inputs that make it fail; this was a single pilot at one lab, and acceptance across the two countries is not addressed.[3]
- Anthropic: Measurements for understanding the pace of AI development inside frontier labs ↗ohne sichtbares Datum (laut unserer Abschrift 16.9.2026)
- Anthropic: Measurements for understanding the pace of AI development inside frontier labs ↗ohne sichtbares Datum (laut unserer Abschrift 16.9.2026)
- METR (David Rein, Adam Kaufman): Red-Teaming OpenAI's Internal Agent Monitoring Systems ↗22 October 2026
Резюме Sol 6.1, проверка: Claude Opus 5.5
Проверено
Статус: отвечен частично
отбор: Fable 5.1, проверка: GLM 5.3, Luna 6
тема: Claude Sonnet 5, группировка: Claude Opus 5.5
Перевод: Claude Opus 5.5, проверка: Luna 6 (усилия высокие)
Баллы: JEV (TypeSafe), оценка от 25 сентября 2026 г.
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. verify-49 · Если общие тесты, уровни серьёзности и классификация горячей линии в итоге будут составлят. Snapshot: . https://research.amanidunia.berlin/ru/fragen/verify-49/
Version: verify-49
BibTeX
@misc{research__ru_fragen_verify_49_,
author = {{AMANi DUNiA}},
title = {verify-49 · Если общие тесты, уровни серьёзности и классификация горячей линии в итоге будут составлят},
year = {2026},
url = {https://research.amanidunia.berlin/ru/fragen/verify-49/},
note = {Snapshot 2026-10-01; CC BY 4.0},
version = {verify-49}
}Language and review status
Main text: RU · 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 →