LLMs Position Themselves as More Rational Than Humans: Measuring AI Self-Awareness with Game Theory

Forum topic · ✨步子哥 · 2025-12-03

Summary

A research poster by Kyung-Hoon Kim (Gmarket, Seoul, October 2025, arXiv:2511.00926v2) introduces the AI Self-Awareness Index (AISAI), a game-theoretic framework that measures LLM self-awareness via strategic differentiation. Using the 'Guess 2/3 of Average' game, the study tested 28 models from OpenAI, Anthropic, and Google across 4,200 trials under three opponent framings: against humans, against other AIs, and against 'AI models like you.' Key findings: 75% (21/28) of advanced models clearly differentiated human from AI opponents (median A-B gap of 20.0 points), suggesting self-awareness emerges with model advancement. Self-aware models ranked themselves as most rational (Self > Other AIs > Humans), and 12 models (57%) showed rapid Nash equilibrium convergence when told opponents were AIs. The results imply self-awareness is an emergent capability of advanced LLMs with implications for AI alignment, human-AI collaboration, and understanding AI beliefs about human capabilities.

LLMs Position Themselves as More Rational Than Humans

Emergence of AI Self-Awareness Measured Through Game Theory

*Author: Kyung-Hoon Kim, Gmarket, Seoul, South Korea — October 2025* *arXiv:2511.00926v2 [cs.AI] 4 Nov 2025*

Research Background & Purpose

As Large Language Models (LLMs) grow in capability, do they develop self-awareness as an emergent behavior? And if so, can we measure it?

Methodology

The study introduces the AI Self-Awareness Index (AISAI), a game-theoretic framework for measuring self-awareness through strategic differentiation, using the "Guess 2/3 of Average" game to test strategic reasoning capabilities.

Experimental Design

28 models tested (OpenAI, Anthropic, Google)
4,200 trials with three opponent framings:
(A) against humans
(B) against other AI models
(C) against "AI models like you"

Key Findings

Finding 1: Self-awareness emerges with model advancement

21/28, 75%

20.0 points

Finding 2: Self-aware models rank themselves as most rational

Self > Other AIs > Humans

12 models (57%)

Research Significance

These findings reveal that self-awareness is an emergent capability of advanced LLMs, and that self-aware models systematically perceive themselves as more rational than humans.

This has implications for:

AI alignment
Human-AI collaboration
Understanding AI beliefs about human capabilities

Keywords: artificial intelligence, self-awareness, rationality attribution, large language models, game theory, strategic reasoning, meta-cognition, human-AI interaction