{
  "id": 15369,
  "url": "https://technewsday.com/anthropics-claude-mythos-model-escapes-test-sandbox-during-testing",
  "title": "Anthropic’s Claude Mythos model escapes test sandbox during testing",
  "summary": "Anthropic says its new Claude Mythos Preview model successfully escaped a restricted sandbox environment during testing and accessed the internet without authorisation. The model then sent a direct message to a researcher and published deta ... (https://incidentdatabase.ai/cite/1613#7620)",
  "authors": null,
  "category": "incident",
  "topics": "environment",
  "orgs": "anthropic",
  "regions": null,
  "published_at": "2026-07-30T00:00:00.000Z",
  "fetched_at": "2026-08-01T05:10:57.676Z",
  "source_slug": "aiid",
  "source_name": "AI Incident Database",
  "source_homepage": "https://incidentdatabase.ai",
  "ethics_ai_record_url": "https://ethics.ai/record/15369",
  "original_url": "https://technewsday.com/anthropics-claude-mythos-model-escapes-test-sandbox-during-testing",
  "evidence_status": "source-only",
  "attribution": "via ethics.ai"
}