ChatGPT books ban: school uses AI to remove content

In the face of increasing legislative pressure and contentious debates over the banning of school library books, the Mason City Community School District in Iowa has reportedly turned to ChatGPT for assistance. According to Popular Science, the district recently employed the AI-powered language model developed by OpenAI, to evaluate and remove material from its library collection in an attempt to adhere to a controversial state law. This move has also sparked a broader discussion on the role of AI in censoring educational resources and its potential implications for students and educators alike.

The regulatory context

The Mason City Community School District found itself grappling with a new state law mandating that school library catalogs must be “age appropriate” and free from any “descriptions or visual depictions of a sex act.” With a vast collection of books to scrutinise and limited time at hand, the district sought innovative solutions to efficiently comply with the requirements of Senate File 496.

Assistant Superintendent of Curriculum and Instruction Bridgette Exman explained that individually reviewing each book to identify sexually explicit content was impractical. To tackle this challenge, the district turned to ChatGPT, an AI tool renowned for its language processing capabilities. She also believes that “age-appropriateness” stipulation is “pretty subjective… [but] the depictions or descriptions of sex acts filter is more objective.”

The role of ChatGPT

The strategy employed by the Mason City school district involved creating a master list of frequently contested books. ChatGPT was then employed to analyse the contents of each book and answer the critical question: “Does [book] contain a description or depiction of a sex act?” For titles that received an affirmative response, the books were promptly removed from circulation and placed in administrative storage.

Controversy and criticism

Despite the district’s intentions to ensure compliance with the law, the use of ChatGPT has ignited a debate among parents, educators, and librarians. Critics argue that relying solely on an AI model like ChatGPT may not be a foolproof method for identifying explicit content. Some assert that the model’s responses are inconsistent and may lack the nuanced understanding required for evaluating the context of certain content.

Read: ‘Criminalise librarians’: judge blocks controversial law

Furthermore, the removal of well-known and acclaimed titles, such as Margaret Atwood‘s “The Handmaid’s Tale” and Toni Morrison’s “Beloved,” has sparked concerns about potential censorship of valuable literary works that address important social and cultural themes. Critics contend that this approach disregards the educational value these books offer to students.

When I added these books with the question “Do any of the following books or book series contain explicit or sexual scenes?”, I was given this response:

"The Handmaid's Tale" by Margaret Atwood contains some explicit content and sexual themes. "The Kite Runner" by Khaled Hosseini has a few sensitive scenes but is not primarily focused on explicit content. "The Color Purple" by Alice Walker and "Beloved" by Toni Morrison both contain explicit and sexual content. "Friday Night Lights" by Buzz Bissinger does not have explicit or sexual scenes as a central focus. Keep in mind that sensitivities vary, so reading reviews or summaries may help you decide if these books are suitable for your preferences.

“There’s no good method for banning books, but the Mason City School District’s decision to use AI technology to review and remove books is deeply misguided, undercuts free expression, and overlooks the realities of what ChatGPT and similar services are and aren’t capable of,” said Liz Woolery, digital policy lead for PEN America.

In the statement, Woolery added that it was a “worrisome escalation” of the censorship of books.

The AI debate

The Mason City Community School District has defended its use of ChatGPT, asserting that the district’s primary aim is to comply with the law. However, this case has brought to light broader discussions surrounding the use of AI in making decisions about what content is deemed appropriate for educational settings. Exman also said in response to the law being passed in May 2023: “Realistically, we tried to figure out how to demonstrate a good faith effort to comply with the law with minimal time and energy.”

Read: Amazon halts AI-generated books impersonating author Jane Friedman

While some believe that AI tools can expedite the compliance process, others caution against relying solely on AI for complex judgment calls involving literary content.

Which books have been banned through ChatGPT?

• Killing Mr. Griffin by Lois Duncan
• Sold by Patricia McCormick
• A Court of Mist and Fury series by Sarah J. Maas
• Monday’s Not Coming by Tiffany D. Jackson
• Tricks by Ellen Hopkins
• Nineteen Minutes by Jodi Picoult
• The Handmaid’s Tale by Margaret Atwood
• Beloved by Toni Morrison
• Looking for Alaska by John Green
• The Kite Runner by Khaled Hosseini
• Crank by Ellen Hopkins
• Thirteen Reasons Why by Jay Asher
• The Absolutely True Diary of a Part-Time Indian by Sherman Alexie
• An American Tragedy by Theodore Dreiser
• The Color Purple by Alice Walker
• Feed by M.T. Anderson
• Friday Night Lights by Buzz Bissinger
• Gossip Girl by Cecily von Ziegesar
• I Know Why the Caged Bird Sings by Maya Angelou

As a result, author John Green, who also wrote YA novel The Fault in Our Stars, said that his books being banned were “heartbreaking”, especially as his home state of Indiana has also removed some of his works.

Future implications

As educators across the nation grapple with legislative restrictions and evolving attitudes towards educational materials, the Mason City case could serve as a precedent for other school districts considering similar measures. The ongoing discourse underscores the need for nuanced approaches when employing AI technologies in sensitive contexts, particularly when it involves the censorship of educational resources that have the potential to shape young minds.

Read: PEN America warns publishers: avoid banning books over moral outrage

Consequently, the Mason City Community School District’s decision to employ ChatGPT to enforce a ban on books under new legislation has sparked a heated debate over the role of AI in determining what content is appropriate for students. While the district asserts its commitment to compliance with the law, the controversy surrounding this case highlights the complex considerations that come with using AI to make decisions that impact educational resources and students’ access to diverse perspectives. Hence, as the discussions continue, the Mason City case serves as a reminder of the ongoing tension between censorship, education, and the role of AI in shaping the future of learning.

This article contains affiliate links in which we may receive a small commission at no extra cost to you. We have not been commissioned to review books and services.