Rising development costs, high rates of clinical trial failures, and growing pressure to provide faster, safer, and more effective treatments all provide hitherto unheard-of difficulties for the pharmaceutical sector. In response, a transforming method to drug development is the combination of big data analytics and generative artificial intelligence (AI). Early phases of drug development are being transformed by generative models including Variational Autoencoders (VAEs), Generative Adversarial Networks (GANs), and transformer-based architectures which enable de novo molecule generation, protein structure prediction, and pharmacokinetic property optimization. Predictive analytics driven by machine learning (ML) and deep learning (DL) methods are improving compound screening, target identification, and clinical trial simulation concurrently. It offers a thorough summary of present approaches, addresses case examples of artificial intelligence-driven discoveries, and assesses the technical support needed to operationalize these developments. While stressing future themes including quantum artificial intelligence, multimodal learning, and AI-driven customized medicine, the study also addresses difficulties including data privacy, model explainability, and validation. In the end, this study shows how generative artificial intelligence might drastically change pharmaceutical innovation when combined with strong data ecosystems.