Jaxsen Day uses ChatGPT to cut through the extra work that can put academic research out of reach for people with visual impairments. Day is a doctoral student in Information Studies at the University of Texas at Austin. He studies accessibility, library systems, visual information, and AI. He came to that work the hard way.
In high school, his mother often had to help the school get materials into forms he could use. When we matriculated as an undergrad at Texas State, that work fell to him. The disability office would send scanned pages, but with no optical character recognition (OCR). He had to learn OCR software and understand where it broke: the text’s reading order scrambled, two columns merged into nonsense, comprehension impossible with garbled text. He fixed it because he had to.
Now Day uses ChatGPT to shorten that path. He built one custom GPT to take a citation and return the UT-specific access link, without forcing him through database interfaces that often hide the right result. He built another to help him explore the paper itself: summarize it by section, answer follow-up questions, and read the summary aloud. For a visually impaired scholar zooming in and using a screen reader, that means he can go straight to the information he needs instead of spending hours making a text usable or skipping through an untagged PDF. He describes that second step as turning a paper into “an instantly accessible version of a document.” With voice mode, he can also revise interview questions while doing his physical therapy or cleaning the kitchen.
That AI-enabled shift can save hours, sometimes days. Without the tools, he says, he could still do the work, but not much else. A search that his custom GPT resolves from a citation would otherwise become a slow hunt through cluttered results, broken clicks, and PDFs that still need remediation before he can decipher them. He now studies the broader consequence of the accessibility burden: what students with disabilities need (but don’t always get) in order to stay in the academic pipeline long enough to finish degrees and enter the workforce. Day also has a paper accepted by the Journal of Academic Librarianship on the barriers academic librarians face when remediating reading materials for students with visual impairments.
At UT Austin, Day is working with his advisor Kenneth Fleischmann as well as campus leadership, including Mario Guerra, to think about how tools like ChatGPT could help more students do their work in greater depth and with less friction. That question is becoming urgent. The Justice Department’s 2024 Title II decision set a more rigorous technical standard for certain web content and mobile apps, with compliance deadlines beginning this April for many educational institutions. With his work, Day is lighting a path to show how AI can enable wider accessibility. In his hands, ChatGPT can turn an inaccessible document into a usable one, and speed the researcher on his way. 
