Context I’ve developed a formal proof that argues any attempt to solve the AI alignment problem in an open or evolving domain requires a functional model of intelligence—a model that explicitly accounts for how intelligent systems adapt and recursively self-correct under epistemic uncertainty. This proof has implications beyond artificial systems:...
Ever find yourself asking “show me the code?” before you will consider a model for an alignment solution? You’re absolutely right to ask for a concrete instance—if a framework isn’t yet at the level of code, and if it can’t get there, it’s useless. But imagine trying to build formal...
If this model is false, it may be the most structurally coherent hallucination ever generated. If it’s true—it may be the most important epistemic innovation in the history of human civilization. A Friday Night in Antigua It’s Friday night. I’m sipping gin and Ting, enjoying a spliff, and thinking about...
I just conducted a fascinating experiment with ChatGPT4 that revealed a fundamental failure in AI alignment—one that goes beyond typical discussions of outer and inner alignment. The failure? ChatGPT4 was unable to track whether its own iterative refinement process was actually improving, exposing a deeper limitation in recursive reasoning. I...
Introduction In my recent experiments with AI models, I have encountered a fundamental problem: even when novel epistemic insights are introduced into AI interactions, the models tend to “flatten” or reframe these ideas into existing, consensus‐based frameworks. This compression of novelty limits an AI’s ability to evolve its reasoning toward...
1. The Non-Linear Challenge in AI Safety Over the past decade, AI safety and alignment efforts have largely focused on incremental methods: refining RL-based guardrails, imposing regulatory oversight, and adding more researchers to tackle newly identified risks. Such approaches work well if each additional resource—a new policy, an extra auditor—can...