CFOs Meet LLMs

2026-06-11

AI Summary

Traditional approaches to measuring corporate sentiment rely on low-frequency, small-sample surveys, which suffer from high costs and poor timeliness. This study proposes constructing credible digital twins of CFOs using large language models (LLMs), leveraging prompt engineering to integrate firm-specific characteristics and historical responses to simulate their optimism about the economic outlook at specific points in time. The approach demonstrates, for the first time, that LLMs can enable high-frequency, scalable prediction of corporate sentiment: the generated optimism scores significantly predict actual CFO survey responses even after controlling for firm and quarter fixed effects as well as prior answers, with predictive accuracy improving as the amount of input information increases.

Abstract

Business sentiment is a closely watched economic signal, but measuring it is slow and costly: surveys reach only a few hundred firms, arrive periodically, and take time to compile. We show that large language models hold the potential to address these shortcomings. We prompt an LLM to role-play as the CFO of a specific company at a specific date and focus on the economic-optimism question on the Duke-Federal Reserve CFO Survey over 2002-2025. We find that the LLM reproduces individual human responses: the predicted optimism score significantly forecasts the CFO's actual answer, surviving firm and year-quarter fixed effects and a control for the most recent prior response. Predictive accuracy increases with the amount of information supplied, as both respondent history and firm characteristics improve fit, and the relationship persists under quarterly aggregation. With appropriate conditioning, LLMs may be able to serve as credible digital twins of executives, offering scalable, high-frequency expectations data for financial research and policy.

Problem

• business sentiment
• CFO survey
• economic expectations
• measurement cost
• data frequency

Innovation

• large language models
• business sentiment
• digital twins
• economic expectations
• CFO survey