The thirty-minute refresh cycle
Why the substrate breathes, and what it means for citizens.
The cadence contract
News scrapers fire every four hours. Wikipedia scanner runs at 03:15 NPT. Ministries KB rebuilds at 04:15. Health substrate re-syncs weekly. Predictive KB rescans daily at dawn. Each source has its own honest cadence — nothing is stamped fresh when it is not.
What the citizen actually notices
A citizen asking about last night's cabinet decision at 08:00 gets the answer. The frozen frontier model cannot help. That single case, multiplied by a million citizens, is the ROI line.
The verifiability payoff
Every NEXUS entry carries a timestamp and a source URL. When the citizen doubts the answer, they click the citation and read the source. Trust is not asserted — it is walkable.
Why we use different words
Where the industry says LLM, we say CLLM. Where it says vector database, we say NEXUS. Where it says Chain-of-Thought, we say PRISM proof-tree. Where it says RAG, we say LATTICE. Where it says trillion-token dataset, we say Smart Router Dataset. Different words because different architecture. Read the whitepaper.