Skip to content
Search
paperAugust 2026Unreviewed

Multi-Agent AI Safety as an Institutional Design Problem

Abdullah X

View resourcePDF

Abstract

AI agents increasingly work inside systems that govern how they delegate tasks, move information, execute actions, and use shared resources. Recent work already shows that deployment rules can change collective behavior. Here we ask which parts of an AI institution produce safety and how they do it. This is the first paper from POLIS, an ongoing research programme studying algorithmic institutions for multi-agent systems. We report a frozen 5,280-episode study suite. The main pre-specified deleg

Categories

Agent Architecture

Cite

@misc{x2026multiagent,
  title = {{Multi-Agent AI Safety as an Institutional Design Problem}},
  author = {Abdullah X},
  year = {2026},
  month = aug,
  eprint = {2608.09828},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2608.09828}
}