Large Language Models Enable Personalized Nudges to Promote Carbon Offsetting Among Air Travellers

2025-08-16

AI Summary

Low voluntary carbon offset uptake and insufficient traveler trust hinder aviation decarbonization. Method: This study pioneers the use of large language models (LLMs) to design personalized, bias-aware behavioral nudges. Leveraging multinational survey data, the LLM infers individual preferences and dynamically generates interventions aligned with users’ cognitive biases and value orientations—emulating human decision-making logic for low-cost, high-precision personalization. Contribution/Results: We extend LLMs beyond generic text generation to structured behavioral intervention design, with a specific focus on trust-building among skeptical users. Empirical evaluation demonstrates a 3–7 percentage-point increase in offset participation rates and an estimated annual reduction of ~2.3 million tons of aviation CO₂ emissions. This work establishes a scalable, AI-driven paradigm for behaviorally informed interventions in transport sustainability.

Technology Category

• Humans and AI: Learning Human Values and Preferences
• Planning, Routing, and Scheduling: Planning with Language Models
• Cognitive Modeling & Cognitive Systems: Simulating Human Behavior

Application Category

• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems

Abstract

Nudge strategies are effective tools for promoting sustainable behaviour, but their impact depends on individual preferences. By emulating human decision-making, large language models (LLMs) offer a cost-effective route for tailoring nudges without extensive behavioural datasets, yet this potential remains unexplored. Focusing on aviation, we use LLMs to design personalized decoy-based nudge strategies that encourage air travellers to voluntarily offset CO$_2$ emissions from flights, and validate their efficacy through 3495 surveys from China, Germany, India, Singapore, and the United States. Results show that LLM-informed personalized nudges are more effective than uniform settings, raising offsetting rates by 3-7$%$ and yielding an additional 2.3 million tonnes of CO$_2$ mitigated annually in aviation. This improvement is driven primarily by increased participation among sceptical travellers with low trust in offset programmes. Our study highlights the potential of LLM-driven personalized nudging strategies for boosting offsetting behaviours to accelerate aviation decarbonization.

Problem

• Personalizing nudges to promote carbon offsetting in air travel
• Using LLMs to tailor nudges without extensive behavioral data
• Validating efficacy of LLM-informed nudges across multiple countries

Innovation

• LLMs tailor personalized nudges for air travellers
• Decoy-based strategies boost CO2 offsetting rates
• LLMs increase participation among sceptical travellers