Collaborative Quest Completion with LLM-driven Non-Player Characters in Minecraft

TL;DR

This study explores how human players collaborate with GPT-4-driven NPCs in Minecraft to complete quests, revealing behavioral patterns and limitations of current language-only AI models in gaming contexts.

Contribution

It introduces a novel Minecraft minigame for studying human-AI collaboration and provides insights into emergent behaviors and current AI limitations in game environments.

Findings

Emergent collaborative behavior patterns identified

Current language models lack rich game-state understanding

Player feedback highlights AI limitations and potential improvements

Abstract

The use of generative AI in video game development is on the rise, and as the conversational and other capabilities of large language models continue to improve, we expect LLM-driven non-player characters (NPCs) to become widely deployed. In this paper, we seek to understand how human players collaborate with LLM-driven NPCs to accomplish in-game goals. We design a minigame within Minecraft where a player works with two GPT4-driven NPCs to complete a quest. We perform a user study in which 28 Minecraft players play this minigame and share their feedback. On analyzing the game logs and recordings, we find that several patterns of collaborative behavior emerge from the NPCs and the human players. We also report on the current limitations of language-only models that do not have rich game-state or visual understanding. We believe that this preliminary study and analysis will inform future game developers on how to better exploit these rapidly improving generative AI models for collaborative roles in games.