Relational Agencies in Children's Writing with Generative AI

While Generative Artificial Intelligence tools that use Large Language Models in programs like ChatGPT are available to the public, little is known about how children learn from and respond to these tools. Much of the previous research with elementary children-the little that exists-has focused on the ability of these tools to produce appropriate text for children to read rather than supporting them in being agents of their own writing. This narrative inquiry explored how a grade 4 child attempted to leverage her agencies to use ChatGPT as a tool to help her write texts that she wanted to write. Implications of this work include the need to understand how children make sense of writing and reading as entangled processes, particularly in the context of Generative Artificial Intelligence agents, and the need to support children in narrating their own agencies when engaging with ChatGPT or other LLMs.