{"type":"rich","version":"1.0","provider_name":"Transistor","provider_url":"https://transistor.fm","author_name":"80,000 Hours Podcast","title":"OpenAI Security: Controlling Models is Now ‘Hell’ (AI Explained cross-post)","html":"<iframe width=\"100%\" height=\"180\" frameborder=\"no\" scrolling=\"no\" seamless src=\"https://share.transistor.fm/e/e89d0228\"></iframe>","width":"100%","height":180,"duration":2411,"description":"This is an (unpaid) cross-post of a video from the podcast 'AI Explained', which Rob Wiblin thought you might be interested in.\nYou can find the original on YouTube here. If you like it and want more similar stuff in future consider subscribing to AI Explained on YouTube, Spotify, Apple or wherever you get podcasts.\n——\n\"This video is hard to summarise. A cracked cipher, an OpenAI security warning, Gemini 4 Argon, RSI paper (co-authored by a who’s who of AI), Lab White House commitments, new hacks emerging, ‘deep personas’, biology Kasparov competitions, and so much more, ending with an epic Opus outro.\"\nIntroduction (00:01:40)\nWrong about Opus 5.5? Deciphering 16th Century Text (00:03:12)\nWhy the models keep breaking out (00:06:41)\nWhat the models aren’t telling us (00:13:20)\nGemini 4 and the race to release (00:17:16)\nWhat happens when AI improves AI? (00:20:59)\nBiology, consciousness, and what we still don’t understand (00:30:11)\nThanks to Philip for permitting the cross-post.","thumbnail_url":"https://img.transistorcdn.com/VO1STE7hN95RRg9QdLo4soV2VhhbR9PF5ZZlRhDYcwE/rs:fill:0:0:1/w:400/h:400/q:60/mb:500000/aHR0cHM6Ly9pbWct/dXBsb2FkLXByb2R1/Y3Rpb24udHJhbnNp/c3Rvci5mbS9zaG93/LzQxNDAyLzE2ODM1/NDQ1NDAtYXJ0d29y/ay5qcGc.webp","thumbnail_width":300,"thumbnail_height":300}