AIvilization v0: Toward Large-Scale Artificial Social Simulation with a Unified Agent Architecture and Adaptive Agent Profiles

2026-02-11

AI Summary

This study addresses the challenge of sustaining long-term autonomy and goal stability among large-scale artificial social agents operating in resource-constrained, dynamic environments. To this end, the authors construct an artificial society integrating a resource-limited sandbox economy with a unified LLM-based agent architecture. They propose a hierarchical branching thought planner, a dual-process adaptive memory system, and a human-in-the-loop intervention mechanism grounded in memory propagation. This framework effectively balances goal stability with environmental responsiveness, successfully reproducing hallmark features of financial markets—such as heavy-tailed returns and volatility clustering—and revealing how disparities in education and resource access drive wealth stratification. The approach demonstrates remarkable robustness in multi-objective, long-horizon tasks.

Technology Category

• Multiagent Systems: Mechanism Design
• Cognitive Modeling & Cognitive Systems: Agent Architectures
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Social Networks and Social Media: Generative AI / large language models and their impact on social systems

Abstract

AIvilization v0 is a publicly deployed large-scale artificial society that couples a resource-constrained sandbox economy with a unified LLM-agent architecture, aiming to sustain long-horizon autonomy while remaining executable under rapidly changing environment. To mitigate the tension between goal stability and reactive correctness, we introduce (i) a hierarchical branch-thinking planner that decomposes life goals into parallel objective branches and uses simulation-guided validation plus tiered re-planning to ensure feasibility; (ii) an adaptive agent profile with dual-process memory that separates short-term execution traces from long-term semantic consolidation, enabling persistent yet evolving identity; and (iii) a human-in-the-loop steering interface that injects long-horizon objectives and short commands at appropriate abstraction levels, with effects propagated through memory rather than brittle prompt overrides. The environment integrates physiological survival costs, non-substitutable multi-tier production, an AMM-based price mechanism, and a gated education-occupation system. Using high-frequency transactions from the platforms mature phase, we find stable markets that reproduce key stylized facts (heavy-tailed returns and volatility clustering) and produce structured wealth stratification driven by education and access constraints. Ablations show simplified planners can match performance on narrow tasks, while the full architecture is more robust under multi-objective, long-horizon settings, supporting delayed investment and sustained exploration.

Problem

• large-scale artificial social simulation
• long-horizon autonomy
• goal stability
• reactive correctness
• adaptive agent profiles

Innovation

• hierarchical branch-thinking planner
• adaptive agent profile
• dual-process memory
• human-in-the-loop steering
• long-horizon autonomy