THE SECURITY OF WORK: PUBLIC PERCEPTIONS OF GENERATIVE AI IN CREATIVE WRITING AND ITS IMPLICATIONS FOR LABOR
Date
2024-05
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
University of Delaware
Abstract
Generative artificial intelligence (GAI) is a type of technology that produces
images, audio, or text generated from the training data, such as information scraped
from the internet. Current innovations in artificial intelligence (AI) have caused many
concerns about AI’s role in reshaping the economy, specifically displacing workers.
There are various discussions on the appropriate policies to regulate AI innovation. In
the background of the Writers Guild of America (WGA) strikes, there is also an
increased need to understand generative AI’s impact on creative labor. This work
examines public opinion on AI-generated content and how that potentially contributes
to the economic security concerns posed by the technology. This project uses an
online human survey (n=359) conducted on Amazon Mechanical Turk using movie
scripts generated by the text-generative interface ChatGPT. Participants assessed the
scripts based on emotional engagement, creativity, and appreciation metrics. To
understand policy implications for technology, the survey asked what policy
frameworks participants believed would safeguard labor rights. This study found that
topical priming affects the participants’ empathy towards the characters in the movie,
and priming, authorship, and complexity affects the participants’ perception of the
economic security concerns.