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

arxiv.org

June 7, 2026

2 min read

🔥🔥🔥🔥🔥

52/100

Summary

Research indicates that large language models (LLMs) are often ascribed human-like attributes such as morality and understanding of natural language. The comparison extends to Age of Empires II, suggesting that both LLMs and the game exhibit similar anthropomorphic characteristics.

Key Takeaways

The study argues that anthropomorphic attributes ascribed to large language models (LLMs) may not be unique and could also apply to other systems, such as video games like Age of Empires II.
The authors propose a 'null' assumption for experiments on LLMs, suggesting researchers should assume non-uniqueness rather than anthropomorphic attributes.
The research emphasizes the need for explicit measurement criteria in discussions about LLMs to avoid circular or uninformative conclusions.
Age of Empires II is demonstrated to be functionally and Turing-complete, supporting the argument about the non-uniqueness of perceived attributes across different substrates.

Read original article

View story →

Coverage & discussions

Hacker News · arxiv.org

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

Community discussions

Hacker News discussion ↗

Historical count: 5 · 5 saved in sample · Historical sample

Analysis ready

Community Sentiment

Negative

Positives

The paper attempts to challenge existing academic methods for evaluating LLM behavior, which could lead to more rigorous standards in AI research.
The discussion around anthropomorphizing LLMs raises important philosophical questions about intelligence and consciousness, pushing the boundaries of AI ethics.

Concerns

Many commenters find the paper's claims incoherent and poorly structured, suggesting it lacks scientific rigor and clarity.
Critics argue that conflating LLMs with human-like attributes is misleading, as it ignores the fundamental differences between AI and human consciousness.