Fusion of art design and technology in contemporary design

Abstract Integrating art, design, and technology has revolutionised the creative process in the modern period, and one tool that has bridged the gap between digital media and traditional artistic methods is Generative Adversarial Networks (GANs). In order to make AI-generated art more realistic and cohesive stylistically, convolutional neural networks use pooling and convolutional layers to notice important artistic details, which are then added to GAN models. This study evaluates GANs for their ability to generate realistic AI-driven artwork using crucial performance metrics such as recall, accuracy, precision, and F1-score. The proposed GAN model outperforms the existing models, CycleGAN, Pix2Pix, and StyleGAN, producing more visually appealing results with enhanced generalisation and classification accuracy. The results demonstrate that artificial intelligence-generated artwork is gradually gaining in popularity and resembling traditional artwork in many ways. This exemplifies the potential revolutionary nature of GANs in the realms of computational design, collaborative art, and digital art production. As AI-driven creativity advances, future research should focus on making things easier to understand, reducing biases in algorithms, and refining the ways in which humans and AI work together to create new things. Performance evaluation measured by classification accuracy, feature similarity, and perceptual quality metrics shows that CNN-assisted GAN models outperform conventional techniques in art style transfer, automated artwork enhancement, and AI-driven creative creation. The research highlights the significance of GAN-based digital art tools in revolutionising contemporary design and expanding the boundaries of digital artistic expression.