Using Games to Learn How Large Language Models Work

📅 2026-03-30
📈 Citations: 0
Influential: 0
📄 PDF
🤖 AI Summary
This work addresses the widespread public gap in intuitive understanding of how large language models (LLMs) operate, compounded by limited AI literacy resources. To bridge this gap, the study introduces gamified learning as a novel pedagogical approach for teaching core LLM mechanisms, presenting two interactive educational games that visually simulate probabilistic modeling and text generation processes. These games illustrate how LLMs predict and sequentially generate words based on their training data. Grounded in established educational principles and thoughtful interaction design, the approach demonstrably enhances learner engagement and conceptual comprehension. Preliminary evaluation indicates that such games hold significant instructional promise for effectively communicating the underlying training and generative dynamics of LLMs, offering an innovative pathway for AI literacy outreach.
📝 Abstract
While artificial intelligence (AI) technology is becoming increasingly popular, its underlying mechanisms tend to remain opaque to most people. To address this gap, the field of AI literacy aims to develop various resources to teach people how AI systems function. Here we contribute to this line of work by proposing two games that demonstrate principles behind how large language models (LLMs) work and use data. The first game, Learn Like an LLM, aims to convey that LLMs are trained to predict sequences of text based on a particular dataset. The second game, Tag-Team Text Generation, focuses on teaching that LLMs generate text one word at a time, using both predicted probabilities of the data and randomness. While the games proposed are still in early stages and would benefit greatly from further discussion, we hope they can contribute to using game-based learning to teach about complex AI systems like LLMs.
Problem

Research questions and friction points this paper is trying to address.

AI literacy
large language models
explainability
public understanding
educational games
Innovation

Methods, ideas, or system contributions that make the work stand out.

game-based learning
large language models
AI literacy
text generation
predictive modeling
🔎 Similar Papers
No similar papers found.
Allison Chen
Allison Chen
Princeton University
I
Isabella Pu
Massachusetts Institute of Technology