Can AI Debias the News? LLM Interventions Improve Cross-Partisan Receptivity but LLMs Overestimate Their Own Effectiveness

TL;DR

This study investigates whether large language models can effectively debias partisan news headlines to improve cross-partisan trust, finding that substantive reframing helps but models overestimate their own effectiveness.

Contribution

The paper demonstrates that LLMs can enhance cross-partisan receptivity through ideological reframing, but they lack accurate self-assessment and psychological fidelity.

Findings

Substantive reframing increased trustworthiness among conservatives.

Lexical debiasing had no significant effect.

LLMs overestimate their intervention effectiveness.

Abstract

Partisan news media erode cross-partisan trust, but large language models (LLMs) offer a potential means of debiasing such content at scale. Across two pre-registered experiments, we tested whether LLM-generated debiasing of liberal news headlines could improve conservative readers' trust-relevant judgments. Study 1 found that subtle lexical debiasing (replacing emotive words with more moderate synonyms) had no effect on any outcome. Study 2 found that a more substantive reframing intervention significantly increased conservatives' perceived trustworthiness, completeness, and willingness to engage with liberal news headlines, without producing a backfire effect among a sample of liberals. In Study 1, the intervention produced robust effects among LLM-simulated silicon participants, whereas it had no impact on human readers. In Study 2, the intervention's effects among silicon participants aligned directionally with human responses but were significantly larger in magnitude for some outcomes. Moderation analyses revealed that the model's implicit theory of who responds to debiasing diverged from the psychological profile that actually predicted human responsiveness. These findings demonstrate that LLM-based debiasing can improve cross-partisan receptivity when targeting ideological framing rather than surface-level language, but that current models lack both the quantitative accuracy and qualitative psychological fidelity to evaluate their own interventions without human oversight.