Internal vs. External: Comparing Deliberation and Evolution for Multi-Agent Constitutional Design

TL;DR

This study compares internal deliberation and external evolution in multi-agent systems across three environments, finding evolution generally outperforms deliberation in collective actions but not in bilateral trading, with performance depending on incentives.

Contribution

First controlled comparison of internal versus external methods for multi-agent constitutional design across diverse social environments.

Findings

Evolution outperforms deliberation in collective-action settings (p < 0.01).

Neither method improves outcomes in bilateral trading.

Evolution's advantage reverses at certain incentive levels, leading to suboptimal cooperation.

Abstract

Multi-agent AI systems need behavioral constitutions, but it is unresolved whether such rules should emerge internally through agent self-governance or be discovered externally through optimization. We present the first controlled comparison of internal deliberation and external evolution across three social environments: a coordination grid-world, an iterated public goods game, and a bilateral trading market. Across 180 simulation runs, evolution significantly outperforms deliberation in collective-action settings (p < 0.01), while neither method improves outcomes in bilateral trading. A multiplier ablation reveals that evolution's advantage inverts when incentives shift: at pool multiplier (m = 0.75) the evolved constitution forces value-destroying cooperation and becomes the worst-performing method. Notably, no deliberation run across thirty trials ever proposed punishment -- the canonical cooperation-sustaining mechanism evolution reliably discovers -- suggesting external optimization wins on peaks while internal self-governance trades peaks for structural responsiveness.