arXiv cs.MA· Sylvain Chassang·· 10 天前AI 评分26
交互式对齐：用演化博弈论研究 AI 智能体群体的长期对齐
Interactive Alignment
AI 导读
论文研究交互式智能体群体（AI、团队、企业、政府）与人类福利的长期对齐问题，构建了一个智能体进行种植、交易与扩张选择的农场博弈，核心对齐选择在于将多少最终产出交给人类、多少投入扩张。研究采用两种互补方法：用宪法描述智能体偏好并由 LLM 解读的 AI 智能体模拟，以及可解析的演化博弈论框架。分析表明，演化博弈论工具能有效近似交互式智能体经济，务实规范执行在维持长期对齐上优于简单利他主义与利他执行。
正文 · 原文
Abstract:This paper is interested in the long-run alignment of populations of interactive agents (in particular AIs, but also teams, firms, and governments) with human welfare. Formally, it studies a farming game in which a population of agents make planting, trading, and expansion choices. The key alignment choice lies in how much final output to send to humans, and how much to invest in expansion. Because human welfare comes at the cost of expansion, this creates evolutionary pressure against alignment. The main question is whether it is possible to set up agents' constitutional principles regarding sharing and trading to ensure that alignment survives in the long run. The paper uses two complementary strategies to investigate the question: an AI-agent simulation where agents' preferences are described by a constitution and interpreted via an LLM; and a tractable analytical evolutionary game theory framework, allowing for rapid and intuitive exploration of the space of agent preferences. The analysis suggests that tools from evolutionary game theory provide a useful approximation of interactive agent economies, and that pragmatic norm enforcement shows promise in maintaining long-term alignment over simpler forms of altruism and altruistic enforcement.
| Comments: | 63 pages, 21 Figures, 5 tables |
| Subjects: | Theoretical Economics (econ.TH); Computer Science and Game Theory (cs.GT); Multiagent Systems (cs.MA) |
| Cite as: | arXiv:2607.25019 [econ.TH] |
| (or arXiv:2607.25019v2 [econ.TH] for this version) | |
| https://doi.org/10.48550/arXiv.2607.25019 arXiv-issued DOI via DataCite |
Submission history
From: Sylvain Chassang [view email]
[v1]
Mon, 27 Jul 2026 19:28:30 UTC (137 KB)
[v2]
Tue, 6 Oct 2026 20:51:09 UTC (210 KB)
来源：arXiv cs.MA · arxiv.org