In an LLM-agent simulation, adding a constitutional charter and an AI mediator reduces a text-based power-preservation score, suggesting institutional design as an alignment lever.
Agentic misalignment: How llms could be insider threats, 2025
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.AI 1years
2025 1verdicts
REJECT 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
Democracy-in-Silico: Institutional Design as Alignment in AI-Governed Polities
In an LLM-agent simulation, adding a constitutional charter and an AI mediator reduces a text-based power-preservation score, suggesting institutional design as an alignment lever.