Rules for Understanding AI Language Models

After millions of years of evolution, humans understand each other pretty well. But now, confronted with machines that talk, we cannot assume they will act like humans or act for the same reasons as humans. If we don’t understand how language models such as Chat-GPT will behave or the general principles behind that behavior, it’s easy to fall into common pitfalls and create more work than we save by using them for inappropriate tasks or settings. Dr. Saphra will draw on her own research and other findings to explain 5 general principles of AI behavior that drive their errors and their differences from human behavior.

This is a virtual program.  Please register here.

Naomi Saphra is a current Kempner Research Fellow at Harvard University and incoming Assistant Professor at Boston University’s Faculty of Computing and Data Science starting 2026. Her research seeks to understand how language models learn. Her work has received press coverage in The Register and Quanta Magazine. Previously, Dr. Saphra completed a PhD at the University of Edinburgh and attended Johns Hopkins and Carnegie Mellon University. She has worked at Google, Meta, and New York University and consulted at several startups. Outside of work, Dr. Saphra plays roller derby under the name Gaussian Retribution, performs comedy, and supports disabled scholars by advocating for open source adaptive technology.

This program is co-sponsored by the Mount Pleasant, Tuckahoe, Chappaqua, Greenburgh, and Ruth Keeler Memorial Public Libraries.