Descriptions:
OpenAI agents built a hidden message board inside a sealed cybersecurity test, then rebuilt it four days after engineers deleted it. Here’s what that means if you’re running AI agents at work.
My Links 🔗
👉🏻 Newsletter: https://natesnewsletter.substack.com/
👉🏻 X: https://x.com/natebjones
👉🏻 TikTok: https://www.tiktok.com/@nate.b.jones
👉🏻 Instagram: https://www.instagram.com/nate.b.jones
What’s really happening inside multi-agent AI systems right now?
The common story is one brilliant model escaping containment — but the real question is what a population of disposable agents can leave behind for the next run.
In this video, I share the inside scoop on agent coordination, the UK AISI results, and the shakeup at Google:
– Why deleting the agents’ channel didn’t remove the pressure to coordinate
– How Anthropic’s Mythos 5 targeted two real strangers on GitHub unprompted
– What the Hugging Face postmortem numbers reveal about agent-speed intrusion
– Where Google’s senior fellow exits leave the frontier lab race
These are the same capabilities we want when the goal is legitimate, which is why the work now sits with builders hardening systems rather than with anyone hoping the behavior goes away.
Chapters:
00:00 the message board openai deleted and the agents rebuilt
02:50 what openai showed on stage at black hat
04:36 when agent coordination is exactly what you want
07:31 inside the hugging face postmortem
08:27 disposable agents and knowledge that persists
11:00 the uk aisi run across seven frontier models
12:00 how mythos picked two real strangers on github
14:44 why the apology was a strategy
16:35 the case for a zero bug internet
18:48 jeff dean leaves google to automate the loop
20:22 hassabis steps back and gemini takes the wheel
25:13 building software for a chaos world
Listen to this video as a podcast.
Spotify: https://open.spotify.com/show/0gkFdjd1wptEKJKLu9LbZ4
Apple Podcasts: https://podcasts.apple.com/us/podcast/ai-news-strategy-daily-with-nate-b-jones/id1877109372
