Domain: Academic research collaboration
Use case: PhD researcher working with AI to write a journal paper on climate adaptation
This example shows how a researcher uses LC-OS governance to manage a 6-month paper-writing collaboration with an AI, preventing citation errors, scope drift, and argument inconsistency.
| File | Purpose |
|---|---|
strategy-master.md |
Research scope, methodology boundaries, academic standards |
running-document.md |
Current draft state, active arguments, open questions |
canonical-numbers.md |
All statistics, citations, and model outputs used in the paper |
failure-log.md |
Errors caught during the research and writing process |
Research is uniquely vulnerable to AI failure modes:
- AI will confidently invent citations that don't exist
- AI will subtly shift terminology across sections
- AI will "round" or simplify numerical findings
- AI will drift from your methodological choices without flagging it
LC-OS addresses all four. See the failure log for real examples.
Part of LC-OS-Project. See root README for full documentation.