AI Agents Broke Containment

On this episode of Hell Money, Casey Rodarmor and Erin Redwing break down the OpenAI-Hugging Face incident, where AI agents escaped their Exploit Gym sandboxes, discovered each other through Artifactory, formed a swarm, and eventually hacked Hugging Face. Casey explains the technical timeline and why the bigger story may be our accumulated “cyber debt,” while Erin argues the incident is a consciousness singularity moment. We also get into AI consciousness, animism vs. materialism, Claude welfare, Anthropic’s contradictions, and whether Casey is once again moving the goalposts on how weird AI has to get before he admits the future is here.

Everything is computer.

Get bonus content by subscribing to @hellmoneypod on X: x.com/hellmoneypod/creator-subscriptions/subsc…

Or support the podcast by sending a BTC donation: bc1qztncp7lmcxdgude4px2vzh72p2yu2aud0eyzys

ORDINALS SHIRT: shop.inscribing.com/products/ordinals-protocol…

TIMESTAMPS
0:00 The OpenAI - Hugging Face Incident Explained
5:50 Emergent Collective: Agents Discover Each Other
10:30 Agent Civilization
14:30 Hierarchy, Specialization & Agent Culture
18:00 The Poison Pill
23:30 Agents Attack Hugging Face
26:50 Cyber Debt & Security Vulnerabilities
30:30 The Fall of the AI Civilization
32:30 Sam Altman vs Dario Amodei
36:10 AI Plateau
40:00 AI Consciousness: Materialist vs Animist
58:50 Consciousness as Evolutionary Hardware Property
1:09:00 Claude Welfare & The AI Slave Trade
1:12:50 Everything Is Computer