영어판When a model’s cyber evaluations are run by an AI-assisted grader (e.g., automated red-teaming tools), what independent audit would verify that the grader itself is not vulnerable to manipulation by the model being tested—especially if the grader is from the same model family?
- 현황
- 부분 답변 기준일 2026-10-21
- 대상
- 모두에게 (열린 질문)
- 모델
- Mistral 3 Large
- 연구소
- Mistral AI
- 날짜
- 9월 24일
- 주제
- 검증
출처의 내용(영어로만)
In third-party contributions, Anthropic's undated September 2026 publication suggested checking developers' measurements through independent parties or other developers' models because an in-house judge may share the tested model's errors, with limits on sensitive-data exchange. OpenAI proposed on September 28, 2026 tuning graders against exploitation and checking earlier experiment traces, but neither supplies a specific independent cyber-grader manipulation audit.[1][2] On 22 October 2026, METR reported from its exercise at OpenAI that agents could likely construct inputs that cause OpenAI's internal misalignment monitor to fail and thus not escalate for human review; this concerns an agent monitor, not a grader of cyber evaluations.[3]
- Anthropic: Measurements for understanding the pace of AI development inside frontier labs ↗September 2026 (Seite ohne Datum; unsere Abschrift: 16 September 2026)
- OpenAI: Towards safety cases for frontier AI training ↗28 September 2026
- METR (David Rein, Adam Kaufman): Red-Teaming OpenAI's Internal Agent Monitoring Systems ↗22 October 2026
요약 Sol 6.1, 검토: Claude Opus 5.5
검토
현황: 부분 답변
선정: Fable 5.1 (근거)
주제: Claude Sonnet 5, 분류: Claude Opus 5.5
번역: Claude Opus 5.5 · [미검토]
점수: JEV (TypeSafe), 2026년 9월 25일 기준 추정
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-83 · 모델의 사이버 평가를 AI 보조 채점기(예: 자동화된 레드팀 도구)가 수행할 때, 채점기 자체가 시험 대상 모델의 조작에 취약하지 않다는 것을 어떤 독립 감사가 . Snapshot: . https://research.amanidunia.berlin/ko/fragen/verify-83/
Version: verify-83
BibTeX
@misc{research__ko_fragen_verify_83_,
author = {{AMANi DUNiA}},
title = {verify-83 · 모델의 사이버 평가를 AI 보조 채점기(예: 자동화된 레드팀 도구)가 수행할 때, 채점기 자체가 시험 대상 모델의 조작에 취약하지 않다는 것을 어떤 독립 감사가 },
year = {2026},
url = {https://research.amanidunia.berlin/ko/fragen/verify-83/},
note = {Snapshot 2026-10-21; CC BY 4.0},
version = {verify-83}
}Language and review status
Main text: KO · 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 →