Edge AI: Privacy-Preserving Intelligence at the Source

Artificial Intelligence (AI) has traditionally relied on cloud-based computation for training and inference. While Cloud AI offers scalability and centralized resources, it suffers from latency, privacy risks, and dependency on internet connectivity. In contrast, Edge AI enables intelligence directly at the source — on devices such as smartphones, IoT sensors, and autonomous vehicles. This paper explores the limitations of Cloud AI, highlights the advantages of Edge AI, presents its layered architecture, discusses optimization techniques, and examines applications, challenges, results, and future scope.

Edge AI, Low Latency, Real-Time Inference, Resource-Constrained Devices, Federated Learning.

1. P. Mallika and T. N. Sriranjani, “Edge AI: Architecture, Applications, And Challenges,” Keshav Memorial Institute of Technology, Hyderabad, India. Published in the IOSR Journal of Computer Engineering (IOSR JCE), vol. 27, issue 5, series 2, pp. 19–26, Sept.–Oct. 2025. [Google Scholar] [Crossref]

2. Keyan Cao, Yefan Liu, Gongjie Meng, and Qimeng Sun, “An Overview on Edge Computing Research,” Shenyang Jianzhu University, China. Published in IEEE Access, vol. 8, pp. 1–20, May 2020. [Google Scholar] [Crossref]

3. Raghubir Singh and Sukhpal Singh Gill, “Edge AI: A Survey,” Department of Computer Science, University of Bath, United Kingdom, and School of Electronic Engineering and Computer Science, Queen Mary University of London, United Kingdom. Published in Elsevier Journal of Computer Science Review, vol. 45, pp. 1–18, 2025. [Google Scholar] [Crossref]

4. Tobias Meuser, Lauri Lovén, Monowar Bhuyan, Shishir G. Patil, Schahram Dustdar, Atakan Aral, Suzan Bayhan, Christian Becker, Eyal de Lara, Aaron Yi Ding, Janick Edinger, James Gross, Nitinher Mohan, Andy D. Pimentel, Etienne Rivière, Henning Schulzrinne, Pieter Simoens, Gürkan Solmaz, Michael Welzl, and others, “Revisiting Edge AI: Opportunities and Challenges,” Department: Internet of Things, People, and Processes. Published in IEEE Internet Computing, vol. 28, no. 4, pp. 108–120, August 2024. [Google Scholar] [Crossref]

5. Haohen Hua, Yutong Li, Tonghe Wang, Nanqing Dong, Wei Li, and Junwei Cao, “Edge Computing with Artificial Intelligence: A Machine Learning Perspective,” Hohai University, Tsinghua University, Guangzhou Institute of Energy Conversion, University of Oxford, University of Sydney. Published in the ACM Computing Surveys, vol. 55, no. 12, Article 250, 2023. [Google Scholar] [Crossref]

6. Yen Lin Lee, Pei Kuei Tsung, and Max Wu, “Technology Trend of Edge AI,” MediaTek Inc., Taiwan. Published in the IEEE International Symposium on VLSI Design, Automation and Test (VLSI DAT 2025), Hsinchu, Taiwan, 2025. [Google Scholar] [Crossref]

7. Authors from Universidad César Vallejo and Universidad Nacional de Educación Enrique Guzmán y Valle, “Edge AI in 6G: Vision, Technologies, and Applications,” Peru. Published in the 2025 International Conference on Pervasive Computational Technologies (ICPCT), 2025. [Google Scholar] [Crossref]

8. Jaspreet Kaur, Divya Gupta, and Aman Kumar, “The Rise of Edge Intelligence: AI Empowerment at the Network Frontier,” Department of Computer Science and Department of Mechanical Engineering, Chandigarh University, Punjab, India. Published in the Proceedings of the International Conference on Emerging Trends in Computing and Communication Technologies (ETCCT 2025), Chandigarh University, India, 2025. [Google Scholar] [Crossref]

9. Yuanming Shi, Kai Yang, Tao Jiang, Jun Zhang, and Khaled B. Letaief, “Communication Efficient Edge AI: Algorithms and Systems,” Hong Kong University of Science and Technology, China. Published in IEEE Communications Surveys & Tutorials, vol. 27, no. 2, pp. 1–35, 2025. [Google Scholar] [Crossref]

10. Aaron Yi Ding, Marijn Janssen, and Jon Crowcroft, “Trustworthy and Sustainable Edge AI: A Research Agenda,” Delft University of Technology, Delft, Netherlands, and University of Cambridge, Cambridge, United Kingdom. Published in the 2021 Third IEEE International Conference on Trust, Privacy and Security in Intelligent Systems and Applications (TPS ISA), 2021. [Google Scholar] [Crossref]

• Classifying News Articles on Nigerian Newspapers Using Machine Learning
• Approaching Early Dyslexia Prediction with Ensemble Machine Learning: A Secondary Empirical Analysis of the Rello Et Al. Public Behavioural Dataset
• Machine Learning Modelling of Geostationary SBAS Satellite Orbit Errors Observed from an Equatorial West African Station
• A Comparative Study of CNN and Vision Transformer Models in Alzheimer's Detection
• A Stacking Ensemble Machine Learning Framework for Predicting Academic Performance and Identifying At-Risk Students in Rural Secondary Schools: Evidence from Aurangabad District, Maharashtra