An Empirical Study of AI Techniques in Mobile Applications
Yinghua Li, Xueqi Dang, Haoye Tian, Tiezhu Sun, Zhijie Wang, Lei Ma, Jacques Klein, Tegawendé F. Bissyandé
TL;DR
The paper addresses the landscape of AI in mobile apps by conducting the largest empirical study to date, covering on-device ML, on-device DL, and cloud-based AI services. It applies an automated AI Discriminator to identify 56,682 real-world AI apps from AndroZoo and analyzes them across three perspectives: application/popularity/update dynamics, framework and model protection, and user privacy and review sentiment. Key contributions include the extensive AI app dataset, detailed findings on deployment trends (e.g., dominance of on-device DL and TensorFlow Lite usage), model protection status, and nuanced user privacy concerns and attitudes, along with practical recommendations for developers, users, and AI R&D. The work provides valuable market insights, highlights security and privacy challenges, and offers a resource for reproducible research and further AI-mobile applications study. The findings have implications for app development, deployment strategies, and privacy-preserving practices in mobile AI ecosystems.
Abstract
The integration of artificial intelligence (AI) into mobile applications has significantly transformed various domains, enhancing user experiences and providing personalized services through advanced machine learning (ML) and deep learning (DL) technologies. AI-driven mobile apps typically refer to applications that leverage ML/DL technologies to perform key tasks such as image recognition and natural language processing. In this paper, we conducted the most extensive empirical study on AI applications, exploring on-device ML apps, on-device DL apps, and AI service-supported (cloud-based) apps. Our study encompasses 56,682 real-world AI applications, focusing on three crucial perspectives: 1) Application analysis, where we analyze the popularity of AI apps and investigate the update states of AI apps; 2) Framework and model analysis, where we analyze AI framework usage and AI model protection; 3) User analysis, where we examine user privacy protection and user review attitudes. Our study has strong implications for AI app developers, users, and AI R\&D. On one hand, our findings highlight the growing trend of AI integration in mobile applications, demonstrating the widespread adoption of various AI frameworks and models. On the other hand, our findings emphasize the need for robust model protection to enhance app security. Additionally, our study highlights the importance of user privacy and presents user attitudes towards the AI technologies utilized in current AI apps. We provide our AI app dataset (currently the most extensive AI app dataset) as an open-source resource for future research on AI technologies utilized in mobile applications.