Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Generative Artificial Intelligence for Internet of Things Computing: A Systematic Survey

Fabrizio Mangione, Claudio Savaglio, Giancarlo Fortino

TL;DR

This survey analyzes how Generative AI can transform IoT computing by surveying literature from 2020 to 2025 using PRISMA. It classifies GenAI methods into explicit and implicit density models and maps their IoT applications across Internet-, Object-, and Semantic-oriented contexts. The study highlights core techniques (e.g., ARMs, VAEs, GANs, diffusion models), deployment challenges on edge devices, and a suite of compression, on-device tuning, and offloading strategies to enable scalable IoT GenAI. It also identifies research gaps, such as hardware-aware model design, secure and privacy-preserving training, and cross-domain integration, and outlines concrete directions for future work with practical impact on real-world IoT ecosystems.

Abstract

The integration of Generative Artificial Intelligence (GenAI) within the Internet of Things (IoT) is garnering considerable interest. This growing attention stems from the continuous evolution and widespread adoption they are both having individually, enough to spontaneously reshape numerous sectors, including Healthcare, Manufacturing, and Smart Cities. Hence, their increasing popularity has catalyzed further extensive research for understanding the potential of the duo GenAI-IoT, how they interplay, and to which extent their synergy can innovate the state-of-the-art in their individual scenarios. However, despite the increasing prominence of GenAI for IoT Computing, much of the existing research remains focused on specific, narrowly scoped applications. This fragmented approach highlights the need for a more comprehensive analysis of the potential, challenges, and implications of GenAI integration within the broader IoT ecosystem. This survey exactly aims to address this gap by providing a holistic overview of the opportunities, issues, and considerations arising from the convergence of these mainstream paradigms. Our contribution is realized through a systematic literature review following the PRISMA methodology. A comparison framework is presented, and well-defined research questions are outlined to comprehensively explore the past, present, and future directions of GenAI integration with IoT Computing, offering valuable insights for both experts and newcomers.

Generative Artificial Intelligence for Internet of Things Computing: A Systematic Survey

TL;DR

This survey analyzes how Generative AI can transform IoT computing by surveying literature from 2020 to 2025 using PRISMA. It classifies GenAI methods into explicit and implicit density models and maps their IoT applications across Internet-, Object-, and Semantic-oriented contexts. The study highlights core techniques (e.g., ARMs, VAEs, GANs, diffusion models), deployment challenges on edge devices, and a suite of compression, on-device tuning, and offloading strategies to enable scalable IoT GenAI. It also identifies research gaps, such as hardware-aware model design, secure and privacy-preserving training, and cross-domain integration, and outlines concrete directions for future work with practical impact on real-world IoT ecosystems.

Abstract

The integration of Generative Artificial Intelligence (GenAI) within the Internet of Things (IoT) is garnering considerable interest. This growing attention stems from the continuous evolution and widespread adoption they are both having individually, enough to spontaneously reshape numerous sectors, including Healthcare, Manufacturing, and Smart Cities. Hence, their increasing popularity has catalyzed further extensive research for understanding the potential of the duo GenAI-IoT, how they interplay, and to which extent their synergy can innovate the state-of-the-art in their individual scenarios. However, despite the increasing prominence of GenAI for IoT Computing, much of the existing research remains focused on specific, narrowly scoped applications. This fragmented approach highlights the need for a more comprehensive analysis of the potential, challenges, and implications of GenAI integration within the broader IoT ecosystem. This survey exactly aims to address this gap by providing a holistic overview of the opportunities, issues, and considerations arising from the convergence of these mainstream paradigms. Our contribution is realized through a systematic literature review following the PRISMA methodology. A comparison framework is presented, and well-defined research questions are outlined to comprehensively explore the past, present, and future directions of GenAI integration with IoT Computing, offering valuable insights for both experts and newcomers.

Paper Structure

This paper contains 29 sections, 3 equations, 5 figures, 13 tables.

Table of Contents

  1. Introduction
  2. Background
  3. Towards Intelligent Internet of Things Computing
  4. Deep Generative Models - DGMs
  5. Taxonomy of DGMs and related IoT-oriented application scenarios
  6. Explicit density
  7. Implicit density
  8. Research Methodology
  9. Objectives
  10. Search Strategy
  11. Eligibility criteria
  12. Study Selection
  13. Literature Review
  14. Internet-oriented perspective
  15. Object-oriented articles
  16. ...and 14 more sections

Figures (5)

  • Figure 1: Classification task with the related decision boundary, inspired from jakub.
  • Figure 2: A taxonomy of DGMs inspired from goodfellow2017nips2016tutorialgenerative.
  • Figure 3: Flow-chart of the literature review selection process according to the PRISMA guidelines.
  • Figure 4: Practical challenges of GenAI for IoT systems.
  • Figure 5: A taxonomy of Model compression techniques.