{
  "id": 6373,
  "url": "https://arxiv.org/abs/2604.03524v1",
  "title": "Structural Rigidity and the 57-Token Predictive Window: A Physical Framework for Inference-Layer Governability in Large Language Models",
  "summary": "Current AI safety relies on behavioral monitoring and post-training alignment, yet empirical measurement shows these approaches produce no detectable pre-commitment signal in a majority of instruction-tuned models tested. We present an energy-based governance framework connecting transformer inference dynamics to constraint-satisfaction models of neural computation, and apply it to a seven-model cohort across five geometric regimes. Using trajectory tension (rho = ||a|| / ||v||), we identify a 5",
  "authors": "Gregory M. Ruddell",
  "category": "research",
  "topics": "regulation,safety-alignment,environment",
  "orgs": null,
  "regions": null,
  "published_at": "2026-04-04T00:08:17.000Z",
  "fetched_at": "2026-07-14T16:32:28.608Z",
  "source_slug": "arxiv-ethics",
  "source_name": "arXiv",
  "source_homepage": "https://arxiv.org",
  "ethics_ai_record_url": "https://ethics.ai/record/6373",
  "original_url": "https://arxiv.org/abs/2604.03524v1",
  "evidence_status": "source-only",
  "attribution": "via ethics.ai"
}