The Evolution of Generative AI: Trends and Applications
Year: 2025
Type: article
Abstract: Generative artificial intelligence (AI) has revolutionized AI by enabling high-fidelity content creation across text, images, audio, and structured data. This survey explores the core methodologies, advancements, applications, and challenges of generative AI, covering key models such as Variational Autoencoders (VAEs), Generative Adversarial Networks (GANs), Diffusion Models, and Transformer-based architectures. These innovations have driven breakthroughs in healthcare, scientific computing, Natural Language Processing (NLP), computer vision, and autonomous systems. Despite its progress, generative AI faces challenges in bias mitigation, interpretability, computational efficiency, and ethical governance, necessitating research into scalable architectures, explainability, and AI safety mechanisms. Integrating Reinforcement Learning (RL), multi-modal learning, and self-supervised techniques enhances controllability and adaptability in generative models. Additionally, as AI reshapes industrial automation, digital media, and scientific discovery, its societal and economic implications demand robust policy frameworks. This survey provides a comprehensive analysis of generative AI’s current state and future directions, highlighting innovations in efficient generative modelling, AI-driven scientific reasoning, adversarial robustness, and ethical deployment. By consolidating theoretical insights and real-world applications, it offers a structured foundation for researchers, industry professionals, and policymakers to navigate the evolving landscape of generative AI.
Source: IEEE Access
Authors: Maria Trigka, Elias Dritsas
Institution: University of West Attica
Language: English
Subfield: Computer Vision and Pattern Recognition
Field: Computer Science
Domain: Physical Sciences
Open Access status: gold
Estimated APC paid: $2,075
Funder: Hellenic Academic Libraries Link
OpenAlex ID: W4410809023