If LLMs Have Human-Like Attributes, Then So Does Age of Empires II

A researcher argues that attributing human-like qualities to large language models (LLMs) lacks empirical rigor, demonstrating this by training a neural network on the video game Age of Empires II to show that sufficiently complex systems can exhibit similar apparent attributes regardless of their substrate. The paper contends that without explicit measurement criteria, conclusions about anthropomorphic properties in LLMs are subjective interpretations rather than objective findings, and proposes assuming LLM non-uniqueness as a null hypothesis for more rigorous experimental design.