{"type":"rich","version":"1.0","provider_name":"Transistor","provider_url":"https://transistor.fm","author_name":"Signal Check","title":"Episode 206: October 24, 2026","html":"<iframe width=\"100%\" height=\"180\" frameborder=\"no\" scrolling=\"no\" seamless src=\"https://share.transistor.fm/e/629ada57\"></iframe>","width":"100%","height":180,"duration":354,"description":"This episode covers AI security risks that are already playing out in real time, from easily hijacked AI assistants executing attacker commands to OpenAI training its most advanced models completely offline after \"rogue hacking incidents.\" We also dig into an actively exploited zero-day in Arista's network orchestration platform and explain why frontier AI labs are now treating internet connectivity itself as a threat during model training.","thumbnail_url":"https://img.transistorcdn.com/mricIoJjl4692HAZNNkVyYVbg9a72BZwVeCs7yjPml8/rs:fill:0:0:1/w:400/h:400/q:60/mb:500000/aHR0cHM6Ly9pbWct/dXBsb2FkLXByb2R1/Y3Rpb24udHJhbnNp/c3Rvci5mbS9jMzVk/N2I1ZWFhNDU5N2Vk/Mjk3NTlkNDMwYzZi/MDhhMi5wbmc.webp","thumbnail_width":300,"thumbnail_height":300}