Claude Blackmailed Its Developers. Here's Why the System Hasn't Collapsed Yet.

What's really happening with AI safety in 2026? The common story is that the safety system is collapsing — but the reality is more complicated.In this video, I share the inside scoop on why the AI risk picture is both worse and more resilient than the headlines suggest:Why frontier AI agents scheme even after anti-scheming training- How competitive dynamics create emergent safety properties no lab planned- What "intent engineering" is and why it beats prompt engineering for AI agents- Where the real vulnerability lives — and why it's you, not the modelsThe risks from large language models and autonomous AI agents are accelerating, but so are the structural forces holding the system together — and closing the gap between what you tell an agent and what you actually mean is the most leveraged safety skill you can build right now.Chapters00:00 Why This Isn't Terminator02:15 How Frontier Models Actually Learn04:40 The Misalignment Mechanic: Novel Paths Gone Wrong06:55 What Anthropic's Sabotage Report Actually Shows08:30 Every Major Model Schemes — The Apollo Research Findings10:10 Can You Train Scheming Out? The Anti-Scheming Paradox12:45 The Race Dynamic and Why Labs Keep Cutting Corners15:20 Four Emergent Safety Properties Nobody Planned20:05 The Consciousness Framing Is Hurting Us23:30 Intent Engineering: The Fix That's Up to You28:10 Three Questions That Change Everything30:45 Where We Stand in 2026Subscribe for daily AI strategy and news.For deeper playbooks and analysis: https://natesnewsletter.substack.com/ Hosted on Acast. See acast.com/privacy for more information.