Hallucination isn't a bug in LLMs — it's the mechanism. The real problem is misallocating non-determinism, and why "SKILLS.md is enough" is exactly backwards.
AI Doesn't Hallucinate. Your Architecture Does.
Related
I Stopped Curating Games. I Built an AI Agent to Run My Portal Instead.
Picture this: You want to build a gaming portal. You need to find the games, playtest them,...
"I Added a /yolo Button to My Local AI Assistant"
My local AI assistant kept asking permission for every tiny step, so I added conversation-level auto-approve without removing the real safety boundaries.
🧩 Runtime Snapshots #17 — BaSS: Browser-as-Shared-Space
BaSS — Browser-as-Shared-Space: what happens when humans and multiple agents share one live, authenticated browser session, and why only one perception architecture can host it.