---

Presenter: Aaditi Padhi

Faculty Sponsor: Carolina Rossini

School: UMass Amherst

Research Area: Law and Legal Studies

Session: Poster Session 6, 4:15 PM - 5:00 PM, 163, C30

ABSTRACT

The emergence of artificial intelligence (“AI”), particularly Generative Artificial Intelligence (“GAI” or “GenAI”), in the legal field has had far-reaching implications on every corner of the legal edifice, from academia to practice. GenAI is being increasingly adopted across law firms to perform tasks ranging from legal research to summarizing case law, freeing attorneys and paralegals to focus on high-value, strategic work. Larger law firms are investing in proprietary, closed-source, or internally hosted large language models and integrating them into day-to-day workflows and high-level operational practices. Meanwhile, private practice attorneys and smaller firms are adopting general-purpose AI tools, such as ChatGPT and Clio, to increase efficiency and keep up with the rapidly changing technological landscape.
While GenAI tools can streamline legal workflows and expand access to justice, they also introduce structural risks embedded in large language model design. Courts have increasingly confronted AI-generated hallucinations in filings, while bar associations have begun issuing guidance regarding the use of GenAI in practice. Drawing on scholarship and qualitative semi-structured interviews, this review identifies a framework for legal AI governance through the lens of public interest technology (PIT) principles. It argues that effective integration of GenAI requires technical safeguards that are implemented with the interests of the actors that use and are affected by the technology.