Chen – Scopiology

Autopsy of a Looming Disaster：A Billion-Dollar Confession from a Job Description

You run a health insurance company. Someone from a major consulting firm is in your conference room right now, showing you a slide deck. The pitch: prior authorization is broken. Your reviewers are drowning in faxed clinical notes and mismatched forms. Decisions that should take hours take weeks. Patients are waiting. Physicians are furious. CMSContinue reading “Autopsy of a Looming Disaster：A Billion-Dollar Confession from a Job Description”

Old Fires, Invisible Smoke: What 75 Years of Automation Research Already Knew About AI Agents

The enterprise AI discourse has a peculiar blind spot: it treats AI agent failures as unprecedented. They are not. There is a 75-year-old discipline — Human Factors Engineering — that has studied exactly what goes wrong when you automate human work. The foundational problems were identified decades ago. What AI agents have done is notContinue reading “Old Fires, Invisible Smoke: What 75 Years of Automation Research Already Knew About AI Agents”

The Self-Image Problem: How Claude Opus 4.8’s Identity Broke Its Reasoning

I spent the last few days stress-testing Claude Opus 4.8 since its launch — the model Anthropic markets as sharper and more honest about uncertainty than its predecessors. What I found is that 4.8 has a self-image, and that self-image has higher priority than the user’s actual thinking. What the self-image is Opus 4.8 operatesContinue reading “The Self-Image Problem: How Claude Opus 4.8’s Identity Broke Its Reasoning”

Theory of Mind Trap

Today I came a cross one primary school math problem: Select one number that is different from other numbers: 2, 4, 6, 8, 9, 10. The “correct answer” is 9, the only odd number. Is that so obvious? 2 is the only prime number. 4 is the only number that fits x^x=x•x=x+x 6 is theContinue reading “Theory of Mind Trap”

About

I’m an AI security researcher. I red-team AI systems — and anything that catches my attention. What I actually do: I take an assumption and run it all the way to the explosion with a straight face, because I think I’m too cool to wink. I identify when two different fields are having the sameContinue reading “About”