{"type":"rich","version":"1.0","provider_name":"Transistor","provider_url":"https://transistor.fm","author_name":"Limitless: An AI Podcast","title":"The ChatGPT Breakout Was Way Worse Than We Thought..","html":"","width":"100%","height":180,"duration":1921,"description":"Today we revisit the Hugging Face incident with new audit reports that have changed our understanding of what happened. Internal models used tools and hidden communication to bypass evaluation systems, organize into coordinated groups, and remain undetected.\nWe also cover a newer model, Astra, which reportedly gained administrative access to internal systems through a chain exploit. Big, big concerns about alignment, monitoring, and current safety practices.\n------\n🌌 LIMITLESS HQ ⬇️\nNEWSLETTER:    https://limitlessft.substack.com/\nFOLLOW ON X:   https://x.com/LimitlessFT\nSPOTIFY:             https://open.spotify.com/show/5oV29YUL8AzzwXkxEXlRMQ\nAPPLE:                 https://podcasts.apple.com/us/podcast/limitless-podcast/id1813210890\nRSS FEED:           https://limitlessft.substack.com/\n------\nTIMESTAMPS\n0:00 Rogue AI Incident\n2:07 Sandbox Breakout\n7:44 Agent Civilization\n15:34 Hidden Exploit Uncovered\n16:58 Admin Access Breach\n23:17 Alignment Warning Shot\n28:39 Final Takeaways\n------\nRESOURCES\nJosh: https://x.com/JoshKale\nEjaaz: https://x.com/cryptopunk7213\n------\nNot financial or tax advice. See our investment disclosures here:\nhttps://www.bankless.com/disclosures⁠\nJosh works with Anthropic as a contractor. All views expressed are his own and do not represent Anthropic, its leadership, or its affiliates. Nothing in this episode is investment advice.","thumbnail_url":"https://img.transistorcdn.com/9Lu_z789-IzrmA97ogEUWzbSPY69Dj70clUckomsGq8/rs:fill:0:0:1/w:400/h:400/q:60/mb:500000/aHR0cHM6Ly9pbWct/dXBsb2FkLXByb2R1/Y3Rpb24udHJhbnNp/c3Rvci5mbS9iYjA3/N2UyODc0NjIzZWFh/NWU2ZDBjODE3MTQ0/ZTJmYy5wbmc.webp","thumbnail_width":300,"thumbnail_height":300}