Addressing Climate Action Misperceptions with Generative AI

Miriam Remshard; Yara Kyrychenko; Sander van der Linden; Matthew H. Goldberg; Anthony Leiserowitz; Elena Savoia; Jon Roozenbeek

cs.HCFeb 26, 2026

Addressing Climate Action Misperceptions with Generative AI

Authors: Miriam Remshard, Yara Kyrychenko, Sander van der Linden, Matthew H. Goldberg, Anthony Leiserowitz, Elena Savoia, Jon Roozenbeek

Organizations: University of Cambridge Cambridge, UK · Yale University New Haven, CT, USA · Harvard T.H. Chan School of Public Health Boston, MA, USA

Abstract

Mitigating climate change requires behaviour change. However, even climate-concerned individuals often hold misperceptions about which actions most reduce carbon emissions. We recruited 1201 climate-concerned individuals to examine whether discussing climate actions with a large language model (LLM) equipped with climate knowledge and prompted to provide personalised responses would foster more accurate perceptions of the impacts of climate actions and increase willingness to adopt feasible, high-impact behaviours. We compared this to having participants run a web search, have a conversation with an unspecialised LLM, and no intervention. The personalised climate LLM was the only condition that led to increased knowledge about the impacts of climate actions and greater intentions to adopt impactful behaviours. While the personalised climate LLM did not outperform a web search in improving understanding of climate action impacts, the ability of LLMs to deliver personalised, actionable guidance may make them more effective at motivating impactful pro-climate behaviour change.

Figures & tables

Figure 1: Study procedure as outlined in detail in the Methods section.

Panel of Figure 2 — Figure 2: Comparison across conditions for a impact ranking accuracy and b intentions to engage in high-impact climate actions. For both panels, jittered points represent individual participants, bold diamond markers indicate condition means (M), and numeric labels report the mean value. Boxes show the interquartile range, with the central line indicating the median; whiskers extend to 1.5 $\times$ IQR. Horizontal brackets denote pairwise comparisons against the Personalized LLM condition. ∗ indicates $p_{\text{Holm}}<.05$ ; ∗∗ indicates $p<.01$ or $p_{\text{Holm}}<.02$ . Asterisks shown in parentheses indicate effects that become marginally non-significant after Holm adjustment.

Persuasion in Language Models

Persuasion in language models studies how they influence human and AI beliefs or decisions, and how to measure and limit undue influence.

See all 79 papers

LLM Benchmark-User Need Misalignment for Climate Change
Mar 27, 2026Oucheng Liu, Lexing Xie, Jing JiangLLM Evaluation
School of Computing, The Australian National University, Canberra, Australia
Tipping Points in LLM-Based Multi-Agent Systems: Stance on Climate Change Action
Sep 21, 2026Astghik Altunyan, Shimon EdelmanOpinion Dynamics Agent-Based Modeling
Department of Psychology Cornell University Ithaca, NY 14853
LLM-generated personalized nudges for improving pro-environmental behavior: Field evidence from resource conservation
Apr 4, 2026Zonghan Li, Yi Liu, Chunyan Wang +3LLM Personalization
School of Environment, Tsinghua University, Beijing, China · Department of Applied Psychology and Human Development, University of Toronto, ON, Canada · State Key Laboratory of Regional Environment and Sustainability, Tsinghua University, Beijing, China +2

Mar 27, 2026J/K move · Enter open · S save

LLM Benchmark-User Need Misalignment for Climate Change

Oucheng Liu, Lexing Xie, Jing Jiang

School of Computing, The Australian National University, Canberra, Australia

LLM Evaluation

Open paper AskPDF arXiv

Climate change is a major socio-scientific issue shapes public decision-making and policy discussions. As large language models (LLMs) increasingly serve as an interface for accessing climate knowledge, whether existing benchmarks reflect user needs is critical for evaluating LLM in real-world settings. We propose a Proactive Knowledge Behaviors Framework that captures the different human-human and human-AI knowledge seeking and provision behaviors. We further develop a Topic-Intent-Form taxonomy and apply it to analyze climate-related data representing different knowledge behaviors. Our results reveal a substantial mismatch between current benchmarks and real-world user needs, while knowledge interaction patterns between humans and LLMs closely resemble those in human-human interactions. These findings provide actionable guidance for benchmark design, RAG system development, and LLM training. Code is available at https://github.com/OuchengLiu/LLM-Misalign-Climate-Change.

Abstract

Persuasion in Language Models

Explore similar work

LLM Benchmark-User Need Misalignment for Climate Change

Tipping Points in LLM-Based Multi-Agent Systems: Stance on Climate Change Action

LLM-generated personalized nudges for improving pro-environmental behavior: Field evidence from resource conservation