{
  "id": 4423,
  "url": "https://arxiv.org/abs/2605.18838v3",
  "title": "Lying Is Just a Phase: The Hidden Alignment Transition in Language Model Scaling",
  "summary": "Scaling laws predict loss from compute but not how capabilities interact. We measure the coupling between reasoning and truthfulness across 63 base models from 16 families and find a regime change invisible to loss curves: below a family-dependent critical scale N_c, capabilities anticorrelate (r = -0.989, p = 4 x 10^{-5} nonparametric permutation test); above it, they cooperate. N_c ~ 3.5B parameters [2.9B, 13.4B] (bootstrap 95% CI), but model size is not the only variable that determines phase",
  "authors": "Adil Amin",
  "category": "research",
  "topics": "safety-alignment",
  "orgs": null,
  "regions": null,
  "published_at": "2026-05-13T03:14:09.000Z",
  "fetched_at": "2026-07-14T16:30:59.237Z",
  "source_slug": "arxiv-ethics",
  "source_name": "arXiv",
  "source_homepage": "https://arxiv.org",
  "ethics_ai_record_url": "https://ethics.ai/record/4423",
  "original_url": "https://arxiv.org/abs/2605.18838v3",
  "evidence_status": "source-only",
  "attribution": "via ethics.ai"
}