Persuasive or Neutral? A Field Experiment on Generative AI in Online Travel Planning

2025-09-12

AI Summary

This study investigates how generative AI’s linguistic tone—enthusiastic, neutral, or toneless—in online travel itinerary planning affects user engagement, purchase behavior, and experience. Employing a large-scale randomized field experiment, we integrate linguistic cue analysis with multidimensional behavioral modeling (e.g., prompt length, click-through rate, subscription conversion rate). Results show that enthusiastic tone significantly increases prompt length; both enthusiastic and neutral tones improve service subscription rates, whereas toneless instructions yield the weakest performance. This work provides the first empirical evidence that AI tone reshapes users’ decision pathways via linguistic framing, establishing linguistic style as a critical design variable for consumer-facing generative AI interfaces. It advances theoretical understanding of human-AI interaction and offers actionable guidelines for designing interpretable, persuasive generative AI interfaces.

Technology Category

• Humans and AI: Intelligent User Interfaces
• Natural Language Processing: Generation
• Planning, Routing, and Scheduling: Planning with Language Models

Application Category

• Social Networks and Social Media: Generative AI / large language models and their impact on social systems
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Economics, Online Markets and Human Computation: Uses of LLMs and GenAI for marketplace design, bidding, and strategic interactions

Abstract

Generative AI (GenAI) offers new opportunities for customer support in online travel agencies, yet little is known about how its design influences user engagement, purchase behavior, and user experience. We report results from a randomized field experiment in online travel itinerary planning, comparing GenAI that expressed (A) positive enthusiasm, (B) neutral expression, and (C) no tone instructions (control). Users in group A wrote significantly longer prompts than those in groups B and C. At the same time, users in groups A and B were more likely to purchase subscriptions of the webservice. We further analyze linguistic cues across experimental groups to explore differences in user experience and explain subscription purchases and affiliate link clicks based on these cues. Our findings provide implications for the design of persuasive and engaging GenAI interfaces in consumer-facing contexts and contribute to understanding how linguistic framing shapes user behavior in AI-mediated decision support.

Problem

• Examining how GenAI's tone affects user engagement
• Analyzing linguistic cues impact on purchase behavior
• Designing persuasive AI interfaces for consumer contexts

Innovation

• Generative AI with positive enthusiasm tone
• Neutral expression AI for user engagement
• Linguistic cues analysis for behavior explanation