Neural Networks within Generative AI: A Review from a Marketing Research Perspective

Abstract

Focusing on the role of neural networks within Generative Artificial Intelligence, this paper reviews their operating principles, recent developments, and implications for research in the marketing discipline. Special emphasis is placed on generative networks that produce synthetic data with a high degree of similarity to original data – such as realistic images, textual, or audio content – thereby expanding the potential for applied marketing practices. Besides significant opportunities opened up by Generative Artificial Intelligence, the review also elaborates on challenges, including technical obstacles like achieving an adequate Nash equilibrium between competing network components, and application-related considerations such as finding reliable metrics to evaluate their business performance. Broader implications are also analysed from recent publications, including ethical dilemmas related to data authenticity. As a result, this study provides a distinct perspective within the intersecting fields of marketing and computer science in the discourse surrounding Generative Artificial Intelligence, underscoring the research innovations arising from neural network models.