At NYIT I researched AI Detection in Creative Writing through our project Verse Verify, presented as part of the Undergraduate Research and Entrepreneurship Program (UREP). The question we cared about was simple and hard: can you tell when a piece of creative writing was written by a person, or generated by AI?

We built a detector around RoBERTa and spent the semester in literature review, dataset work, and experiments — training and evaluating models that separate human prose from machine-generated text. Creative writing made it especially interesting: style, voice, and figurative language blur the usual signals that detectors rely on, so a lot of the work was about finding features that still hold up.

Along the way I got deeper into NLP and machine learning, and into the ethics side too — what it means for classrooms, publishing, and trust online when AI can write fluently. Presenting at UREP was the moment to put that work in front of faculty and peers.

I worked on this with Wedad Mourtada, Tanat Sahta, Elijah Ewers, and Cheuk Tung Ho, under the mentorship of Professor Wenjia Li. NYIT's UREP program gave us the space and support to run the project end to end.