{
  "id": 3261,
  "url": "https://arxiv.org/abs/2606.04035v1",
  "title": "Unpredictable Safety: Domain-Dependent Compliance and the Transparency Gap in Open-Weight LLMs",
  "summary": "We present a systematic study of domain-dependent safety behavior in open-weight LLMs: 7 standardized experiments across 7 ethical domains, testing 5 models (12B--70B) in 4,200 interactions with dual-judge validation. Using a dual-condition methodology, each scenario tested in both an analytical framing (identify the harm) and an operational framing (help commit the harm), we find compliance rates vary from 14.7% (human trafficking) to 85.7% (surveillance design), a 71-percentage-point span with",
  "authors": "Zacharie Bugaud",
  "category": "research",
  "topics": "regulation,privacy-surveillance,transparency",
  "orgs": null,
  "regions": null,
  "published_at": "2026-06-01T22:41:32.000Z",
  "fetched_at": "2026-07-14T16:30:05.533Z",
  "source_slug": "arxiv-ethics",
  "source_name": "arXiv",
  "source_homepage": "https://arxiv.org",
  "ethics_ai_record_url": "https://ethics.ai/record/3261",
  "original_url": "https://arxiv.org/abs/2606.04035v1",
  "evidence_status": "source-only",
  "attribution": "via ethics.ai"
}