Evaluating the Persuasive Influence of Political Microtargeting with Large Language Models

Advances in LLMs have raised concerns over scalable, personalized political persuasion. In this talk, based on a paper recently published in PNAS, we integrate user data into GPT-4 prompts in real-time, facilitating the live creation of messages tailored to persuade individual users on political issues. We then deploy this application at scale to test whether personalized, microtargeted messaging offers a persuasive advantage compared to nontargeted messaging. We find that while messages generated by GPT-4 were persuasive, in aggregate, the persuasive impact of microtargeted messages was not statistically different from that of nontargeted messages. These findings suggest—contrary to widespread speculation—that the influence of current LLMs may reside not in their ability to tailor messages to individuals but rather in the persuasiveness of their generic, nontargeted messages.

Please join the event. Everyone is welcome—it is free and you do not need to be affiliated with the university.

About Kobi Hackenburg

Kobi Hackenburg is a PhD candidate in Social Data Science at the Oxford Internet Institute, University of Oxford. His doctoral research, funded by a Clarendon Scholarship and supervised by Helen Margetts and Scott Hale, investigates the persuasive influence of personalized AI systems. More broadly, his work lies at the intersection of computation, language, and society. Alongside his PhD, he works as a Doctoral Researcher in the Public Policy Programme at The Alan Turing Institute, the UK’s national institute for AI and data science.