AI Futures at FASS — Brown Bag Session #2 | Dr Morar Florin-Stefan and Prof Lorenz Goette

Dear all,

You are cordially invited to the second session of the AI Futures at FASS — Brown Bag Sessions series in Semester 1 of AY26/27. Speakers will share one result, challenge, question or provocation about AI in relation to their work. Dr Morar Florin-Stefan (NUS History) and Prof Lorenz Goette (NUS Economics) will be presenting on "Learning from a Snowflake: Objectivity, Trained Judgment, and the Promise of AI for Historical Research" and "The Miscalibration of LLM Judgments" respectively.

AI Futures at FASS — Brown Bag Session #2

Date: 17 September 2026, 12pm-1pm
Venue: In-person (FASS Research Division Seminar Room AS7 06-42) and Zoom
Register: In person or Zoom
Lunch will be provided for in-person attendees.

Talk #1: "Learning from a Snowflake: Objectivity, Trained Judgment, and the Promise of AI for Historical Research" (Dr Morar Florin-Stefan)

This talk asks how AI can pose new questions and answer old ones within historical inquiry. Revisiting ideals of objectivity and trained judgment, it examines early results from two computational projects to discuss the role of evidence, interpretation, and expertise. The talk hopes to be a reflection on how historians might use AI critically without sacrificing practices that make history meaningful.

Talk #2: "The Miscalibration of LLM Judgments" (Prof Lorenz Goette)

LLMs are increasingly used to assist in decision making. We show that LLMs suffer from miscalibrated judgments in several domains. With regard to their reasoning, LLMs are overconfident, far more so than their human counterparts. This overconfidence spills over to users. In forecasting tasks, by contrast, LLMs are underconfident and use additional information inefficiently.

Florin-Stefan Morar is Assistant Professor in the Department of History at the National University of Singapore and Director of the NUS History+AI Lab. A historian of science and technology, he studies global knowledge exchange, computational humanities, and AI in historical research. He received his PhD from Harvard University.

Lorenz Goette is a behavioral economist at the National University of Singapore. His research examines decision-making, behavioral biases, and their implications for policy. He studies biases in large language models and how AI shapes human decisions. He is Principal Investigator of the SSRC project “debiaSING” on biases in artificial intelligence.