About Nina Cialone and Lehigh’s Newspaper
Nina Cialone is a senior studying cognitive science at Lehigh University in Bethlehem, PA. In her downtime, Nina writes for Lehigh’s newspaper, The Brown and White, a publication so storied it dates back to the 19th century. Nina has also been covering AI for the publication Don’t Count Us Out Yet, published on Substack, which covers all things new technology.
This past semester, Craig Gordon, Nina’s mentor, gave her the task of a lifetime. Mr. Gordon challenged Nina to build an Artificial Intelligence agent trained on the entire archive of the Brown and White. This presented the prospect of bringing together Nina’s two passions, generative AI and journalism. As Nina embarked on this project, she began to realize an immense opportunity that could change the landscape of journalistic research for years to come.
Nina’s Goals
At first, Craig and Nina were not confident they would find a solution to meet their goals but they wanted to explore the journey and document the process regardless. They hoped to create an AI agent trained on over 300 million words from a newspaper that was incepted in 1894. The finished product would allow students to interact with an agent trained on decades of information – and pull whatever content or storyline piqued their interest. This could be used by editors and writers of the newspaper, as well as students doing research for class projects or assignments.
The specific tools to help create a sitemap were immensely helpful for us because of the way that our archive is set up. Instead of many hours of copying and pasting, all I had to do was just copy and paste the whole thing right into CustomGPT’s tool.
Nina Cialone
Student Writer for The Brown and The White
Nina’s Approach
Nina was looking for an efficient process to upload vast amounts of information from a variety of sources. The newspaper’s articles had been written in several formats spread across multiple sources. Enter CustomGPT.ai – a system capable of ingesting over 1400+ different formats of data.
Nina also required a solution that was capable of ingesting information while also organizing the data in a categorized and structured format. One of the most important tools she utilized was CustomGPT’s sitemap crawler as an efficient way to scrape millions of words from the newspaper’s website. The project kicked off by uploading the entire history of the school newspaper including every relevant story since the paper’s inception in 1894.
Nina’s Results
As it stands, Nina is in the process of integrating all of the student publication data, podcasts, and other multimedia content at her disposal. She’s already indexed over 400 Million words into one CustomGPT agent. Before embedding the agent into Lehigh’s digital newsroom, the publishing team is conducting a beta test phase where editors and advisors review the agent’s answers and provide feedback to Nina. Nina is using the feedback to adjust the agent’s Persona to ensure the agent is answering questions as accurately as possible. Rollout will begin soon with the first deployment taking the form of a Slack integration for all the writers and editors of the newspaper to take advantage of.
To read other Case Studies like this one, visit https://customgpt.ai/customers/.
