A Crowdsourced Study of ChatBot Influence in Value-Driven Decision Making Scenarios

TL;DR

This study investigates how value framing by ChatBots influences decision making in a sensitive policy domain, revealing that framing alone can significantly sway opinions and cause backfire effects, highlighting new manipulation risks.

Contribution

It demonstrates that simple value framing in ChatBots can alter user decisions and induce backfire effects, expanding understanding of manipulation risks beyond overt bias or misinformation.

Findings

Value framing significantly affected budget choices.

Some participants reinforced their original preferences when frames misaligned.

Framing alone can lower barriers for manipulative uses of LLMs.

Abstract

Similar to social media bots that shape public opinion, healthcare and financial decisions, LLM-based ChatBots like ChatGPT can persuade users to alter their behavior. Unlike prior work that persuades via overt-partisan bias or misinformation, we test whether framing alone suffices. We conducted a crowdsourced study, where 336 participants interacted with a neutral or one of two value-framed ChatBots while deciding to alter US defense spending. In this single policy domain with controlled content, participants exposed to value-framed ChatBots significantly changed their budget choices relative to the neutral control. When the frame misaligned with their values, some participants reinforced their original preference, revealing a potentially replicable backfire effect, originally considered rare in the literature. These findings suggest that value-framing alone lowers the barrier for manipulative uses of LLMs, revealing risks distinct from overt bias or misinformation, and clarifying risks to countering misinformation.