Using Markov Chain Text Generators To Facilitate Found Poetry Creation

Summary This is a flexible educational module that simultaneously teaches students about AI techniques and Found Poetry techniques. It asks students to reflect on both the nature of authorship in human-AI creative interaction, and the effect of AI on humans' creative process. It also introduces a conversation about the ethics of AI enterring the art realm. The module can be used as either an outreach or in-class activity, or as a traditional homework assignment.
Topics Markov Chains, Human-AI Interaction, AI and Society
Audience Poetry undergraduates (although see Variants)
Difficulty Introductory 30 minute activity (although see Variants)
Strengths The activity achieves multidisciplinary LOs and can be easily modified for different audiences and settings.
Weaknesses The efficacy of the activity in achieving LOs has not been formally statistically verified.
Dependencies Coding is done in Python 3 and can be performed in a web browser without any software installation.
Variants We include guidelines for tailoring this activity to both K-12 students, CS undergraduates, Poetry undergraduates, and for deploying the activity as an asynchronous assignment rather than synchronous learning activity. Elaboration on tailoring the assignment is included in the Instructor Guide below.

Materials

Instructor Guide

Core Material

Blackout Poetry Presentation
Blackout Poetry Presentation with Notes
Student Handout
Text Files

Supplemental Material

Markov Chain Presentation
Markov Chain Presentation with Notes
Theory of Blackout Poetry Presentation
Theory of Blackout Poetry Presentation with Notes
CS Student Handout
Poetry Student Handout

Examples

Generated text_files