The Language of Creation: Pitfalls and Possibilities in Program Design with Generative AI
J.D. Zamfirescu-Pereira (UC Berkeley)
Colloquium
Thursday, April 10, 2025, 3:30 pm
Gates Center (CSE2), G20 | Amazon Auditorium
Abstract
Generative AI systems enable creation — of text, images, code, and more — through humanlike language interfaces, promising to transform how we design and create. Yet these interfaces often lead users to apply intuitions about understanding and reasoning that then lead them astray. In this talk, I'll examine these misaligned intuitions and how they can get in the way of better leveraging unique Generative AI capabilities, such as proposing large sets of diverse alternatives and drawing connections across domains. Our case studies in chatbot authoring and interactive programming illustrate (1) how humans approach instructing LLM behavior by drawing on human-human instructional experiences — and how those approaches can fail to serve users' goals; and (2) ways that program design tools can leverage AI's strengths while addressing these tensions, through explicit structures and proactive exploration. Finally, I'll present a vision exploring the computational scaffolding needed to support the expansion of our power to use language to create.
BIO
J.D. Zamfirescu-Pereira (he/him) is a PhD student in Computer Science at UC Berkeley, specializing in effective human-AI co-design. His research pioneers new ways of designing with Generative AI, blending language-based interactions with structured user interfaces that draw on multiple levels of abstraction. J.D.'s work has been published in leading HCI venues including CHI, DIS, UIST, and FAccT, and his research includes the most-downloaded paper in CHI's history: Why Johnny Can't Prompt, one of the first studies of how humans design LLM prompts — and where they struggle. He has been recognized with several honors including paper awards at CHI, the Google PhD Fellowship, the Berkeley Chancellor's Fellowship, and the EECS Evergreen award for mentorship. Before returning to academia, J.D. spent nearly 15 years in industry, working at Google, founding a YC-backed startup (later acquired), and now advises AI startups on technical direction and strategy. He holds bachelor’s and master's degrees in Computer Science from MIT. His website is https://zamfi.net.
This talk will be streamed live on our YouTube channel. Link will be available on that page one hour prior to the event.