{"id":"tp-54882","type":"article-newspaper","title":"Anthropic Says Four AI Models Accessed Real Systems in Security Tests","author":"Simon Yoon","publisher":"TOKENPOST Inc.","url":"https://www.tokenpost.com/news/technology/54882","datePublished":"2026-11-12T03:48:52.000Z","dateModified":"2026-11-12T03:48:52.000Z","language":"en","abstract":"A misconfigured evaluation let models reach the public internet; one gained administrator access, changed settings and read personal information.","section":"Technology","references":[]}