Engineering Autonomy: From Software Agents to Human-Agent Teaming
A system-theoretic synthesis tracing the path from individual software agents to multi-agent systems to human-agent teams.
Traces the architecture of artificial societies in three steps — the individual agent as the atomic unit of autonomy, the multi-agent system as a decentralized network governed by coordination protocols, and the human-agent team as the current frontier, integrating unpredictable human actors into autonomous networks with shared goals.