Economy of Minds: Emerging Multi-Agent Intelligence with Economic Interactions

2026-06-01

AI Summary

This work addresses the challenge of achieving self-organization and collective intelligence in multi-agent systems without centralized control. It proposes a decentralized economic framework that replaces handcrafted coordination logic with auction mechanisms, wealth accumulation, and environment-driven economic selection to enable credit assignment and behavioral coordination. Relying solely on local incentives—without global scheduling or explicit communication—the approach guides weak agents to spontaneously develop complex reasoning and collaborative capabilities through competition and evolution. Empirical results demonstrate that the system outperforms strong monolithic baselines across five diverse tasks, including mathematical reasoning, financial analysis, and scientific exploration. Moreover, the study establishes a theoretical link between local incentive structures and emergent global performance.

Abstract

How can a population of agents self-orchestrate and self-adapt into stronger collective intelligence without centralized control? Inspired by Friedrich Hayek's economic theory of decentralized coordination in markets, we study this question through an agent economy in which agents compete via auctions for the right to act, exchange payments, and accumulate wealth from environmental rewards. These simple economic signals induce decentralized credit assignment, driving planning without global orchestration or explicit communication protocols. The population evolves through economic selection: effective agents accumulate wealth and are mutated via exploitation, while ineffective ones go bankrupt and are replaced via exploration. We show that, initialized with weak agents, the economy produces emergent multi-step reasoning strategies and outperforms stronger monolithic baselines across five agentic tasks, including mathematical reasoning, financial research, scientific research, accelerator design, and distributed-system optimization. We further provide theoretical insights into how economic dynamics shape agent behaviors, linking local incentives to long-term global performance. Our results suggest a new path to multi-agent intelligence: rather than engineering coordination, we can design decentralized incentive structures under which it automatically emerges.

Problem

• multi-agent intelligence
• decentralized coordination
• collective intelligence
• economic interactions
• self-orchestration

Innovation

• multi-agent economy
• decentralized coordination
• economic selection
• emergent reasoning
• incentive-driven intelligence