How generative AI companions alleviate situational vulnerability and enhance consumer well-being
Ahmed Gamal Elkattan et al.
Abstract
Purpose Generative AI-enabled chatbots are now capable of engaging in natural, dynamic and relationship-oriented interactions. Building on strength-based approach and transformative service research, this study explores consumer experience with GenAI companions apps designed to provide consumers with synthetic interaction partners. It examines how such companions can be effective in alleviating situational vulnerability (e.g. depression, loneliness) and in supporting sustained well-being. Design/methodology/approach Using large-scale data from GenAI companion apps and online communities, this study employs a multi-method approach that combines fine-tuned large language models (LLMs) to detect situational vulnerability, BERTopic modelling to identify service experience dimensions, and debiased Lasso regression to estimate their effects on situational vulnerability alleviation and consumer well-being. Findings The findings indicate that GenAI companions can meaningfully alleviate situational vulnerability through key service experience dimensions such as AI-enabled coping strategies, humorous interactions, AI intelligence and emotional bonding. However, design frictions such as app accessibility issues, AI sexualisation and content censorship, and communication and memory challenges, may hinder their effectiveness. Additionally, the results confirm the mediating role of situational vulnerability alleviation that mediates the relationship between service dimensions and influencing app recommendations. Originality/value This study offers one of the first empirical, theory-driven examinations of GenAI companions as strength and transformative well-being services. By applying the strength-based approach within a Transformative Service Research framework, it advances understanding of how GenAI companions can function as vulnerability-alleviating service mechanisms and provides actionable insights for the design and governance of AI-enabled well-being services.
Evidence weight
Balanced mode · F 0.40 / M 0.15 / V 0.05 / R 0.40
| F · citation impact | 0.50 × 0.4 = 0.20 |
| M · momentum | 0.50 × 0.15 = 0.07 |
| V · venue signal | 0.50 × 0.05 = 0.03 |
| R · text relevance † | 0.50 × 0.4 = 0.20 |
† Text relevance is estimated at 0.50 on the detail page — for your query’s actual relevance score, open this paper from a search result.