Concept
The Ch 11 salvation case, structurally specified by Ch 12. Faithful LLMs running with institution-owned substrates — corpus, objective, deployment all governed by the institution rather than by a commercial vendor — function as a capability extender for preserve-and-retrain: decompression-on-demand against the institution’s preservation archive for trainees, Socratic instruction at scale, drafting and synthesis that versatile experts then verify. Not a replacement for institutional experts but a tool that makes their work go faster and reach more receivers. Requires open weights, transparent corpus custody, credibly neutral objective specification, and institution-controlled deployment — none of which is the default in the commercial LLM landscape. The integration-friendly LLM configuration is technically realizable, economically marginal, and politically blocked by the same forces that captured the social-media equilibrium.
Connections
- Defined in: Infrastructure for Integration
- Discussed in: Ch 11 — AI as a New Kind of Node, Ch 12 — Infrastructure for Integration
- Enabled by: Decompression on demand
This is a graph landing page. Use the local graph and backlinks (right) to explore, or open the interactive ontology.