Language Games as the Pathway to Artificial Superhuman Intelligence

TL;DR

This paper proposes using language games with role fluidity, reward variety, and rule plasticity to break the data reproduction trap in large language models, enabling open-ended exploration and progression toward artificial superhuman intelligence.

Contribution

It introduces a novel framework of language games that promotes diversity, novelty, and continual evolution in data generation for AI development.

Findings

Language games increase data diversity and coverage.

Multi-agent role fluidity fosters exploration.

Ecosystem scaling enables unbounded data streams.

Abstract

The evolution of large language models (LLMs) toward artificial superhuman intelligence (ASI) hinges on data reproduction, a cyclical process in which models generate, curate and retrain on novel data to refine capabilities. Current methods, however, risk getting stuck in a data reproduction trap: optimizing outputs within fixed human-generated distributions in a closed loop leads to stagnation, as models merely recombine existing knowledge rather than explore new frontiers. In this paper, we propose language games as a pathway to expanded data reproduction, breaking this cycle through three mechanisms: (1) \textit{role fluidity}, which enhances data diversity and coverage by enabling multi-agent systems to dynamically shift roles across tasks; (2) \textit{reward variety}, embedding multiple feedback criteria that can drive complex intelligent behaviors; and (3) \textit{rule plasticity}, iteratively evolving interaction constraints to foster learnability, thereby injecting continual novelty. By scaling language games into global sociotechnical ecosystems, human-AI co-evolution generates unbounded data streams that drive open-ended exploration. This framework redefines data reproduction not as a closed loop but as an engine for superhuman intelligence.