Win Zuniga

This zine explores a dataset collected for the purposes of AI training. The data is sourced from a dataset available on Hugging Face, an online platform and community for artificial intelligence and machine learning. The data consists of captions written by humans. The data was collected through "crowd-sourcing," which is the practice of getting data, labor, ideas, or problem-solving help from a large, online group of people instead of traditional employees. These people are typically given a small amount of money for completing the task, and many of them do this to supplement their income. This investigation calls into question the quality of the data, asking the viewer to think about what could lead to these types of answers.