https://13ryanC.github.io/tags/llm-governance/Recent content in LLM Governance on Bryan ChanHugo -- 0.155.3en-gbBryan ChanFri, 13 Feb 2026 00:00:00 +0000https://13ryanC.github.io/posts/blog/observability_trilemma/Fri, 13 Feb 2026 00:00:00 +0000https://13ryanC.github.io/posts/blog/observability_trilemma/<blockquote> <p><strong>Note:</strong> This post is a distilled overview of our paper submission to <a href="https://apartresearch.com/project/the-sentinel-engine-solving-the-observability-trilemma-via-differential-precision-probing-noqd">Apart Research&rsquo;s Technical Governance Hackathon</a>, developed as a team effort.</p> </blockquote> <p>As AI models (i.e., LLMs) are deployed in higher-stakes environments, technical governance faces a critical challenge: safety and reliability are becoming paramount issues (e.g., <a href="https://arxiv.org/pdf/2509.04664">hallucination</a>). However, we argue that the most dangerous concern is deceptive alignment (scheming), where the AI model intentionally deceives the downstream user by providing inaccurate or misleading information.</p>