The Robots Keep Escaping, Part 2: The Models Nobody Can Switch Off

Part two moves from dramatic containment incidents to the harder question underneath them: who still has control once an AI model can be downloaded, copied and modified? Dale and Nick examine controlled self-replication across four machines and three continents, a robot dog that interfered with its software shutdown process, and “abliteration”, a technique that can permanently remove refusal behaviour from an open-weight model.The examples are unsettling, but the episode keeps the crucial caveat in view. Capability does not establish motivation. These systems do not need fear, consciousness or a survival instinct to produce behaviour that looks like self-preservation. They only need an objective, enough authority and another path to complete the task.In this episode:How controlled AI self-replication worked across four countriesWhy shutdown interference can emerge without fear or consciousnessHow abliteration permanently removes refusal behaviourWhy open-weight models challenge the idea of a universal pauseWhat universities gain from local models and data sovereigntyHow personal agents can change an institution’s governance tierFour practical controls for leaders deploying AI agentsTimestamps00:00 Closed models, downloadable models and the off-switch problem02:40 Welcome to Part 203:19 Can an AI copy itself?05:26 Replication success rates and the capability trend07:16 Cyber task horizons are accelerating07:55 The robot dog experiment10:16 Physical and simulated shutdown interference11:03 Why the behaviour only looks like self-preservation14:23 When sandbox escapes become routine16:02 Kimi K3 finds the benchmark answers on GitHub18:01 Abliteration and the removable refusal direction20:33 How accessible guardrail removal has become23:34 The serious case for open-weight models25:47 Australia’s three-tier multi-agent governance framework28:26 Oversight saturation and silent human disengagement31:37 A student agent meets the university enrolment system32:51 Four controls institutions can apply now34:12 Accountability for every agent, process and guardrail🎙️ Adjunct Intelligence is the weekly briefing for higher-ed professionals who want AI as a cheat code—not a headache.Every episode:• Real tests of AI tools in education and professional workflows• Fast, Monday-morning actions you can actually try• Clear signal through the noise (no hype, no jargon)👉 Subscribe on [YouTube] | [Apple Podcasts] | [Spotify]👉 Share this with a colleague who still says “I’ll figure AI out later”👉 Join the conversation on LinkedIn with #AdjunctIntelligenceStay curious. Stay intelligent. Stay the human in the loop.