When an AI Account Becomes a Security Principal
A practical account-security problem leads to a precise Zero Trust lesson: phishing-resistant sign-in is one boundary, not the whole authorization architecture for an AI account.
> I'm William. Here, I archive my observations on software architecture and AI-native systems, blending corporate leadership experience with a quantitative mindset.
A practical account-security problem leads to a precise Zero Trust lesson: phishing-resistant sign-in is one boundary, not the whole authorization architecture for an AI account.
A first-print field note on choosing CAD, sculpting, or surface modeling from the geometry and constraints of the object.
An evidence-first framework for reading software architecture through eight dimensions, three pressures, quality goals, and observable system behavior.
An architecture-first analysis of OpenDroid, its planner–action–observation loop, Accessibility Service, AppFunctions, and the reliability, permission, and blast-radius trade-offs of mobile agents.
AI does not love or wonder like a human, but a forbidden concept still leaves a trace in the prompt and can pull generation toward the very thing we wanted to avoid.
A systems-level analysis of bi-temporal data architecture for trustworthy AI, focused on accuracy, latency, provenance, and operational trade-offs.
A system-architecture reading of AI skills, separating prompt text, workflow logic, and MCP-based tool execution so teams can reason about latency, control, and blast radius.
OpenClaw hype vs. reality: autonomous agents as glue layers, Copilot vs. mercenary metaphors, and why full local access or blind VPS quarantine both miss a sound architecture.