AI-Driven Drug Discovery Pipelines: A Systematic Literature Review on Advances, Challenges, and Future Directions

The swift adoption of artificial intelligence (AI) in drug discovery has reshaped conventional processes, creating new possibilities to hasten and refine the detection and creation of therapeutic compounds. This systematic literature review investigates the present landscape of AI-driven drug discovery, with particular attention to its applications in varied fields including general drug development, antibiotic and antimicrobial research, cancer therapeutics, peptide and antibody design, drug repurposing, and Traditional Chinese Medicine. We rigorously examine the progress, obstacles, and prospective trajectories of AI in these domains, while identifying both the technological breakthroughs and the enduring limitations that impede broad implementation. The review consolidates results from an extensive range of studies by applying a systematic approach to discern primary patterns, research techniques, and conclusions. The study shows artificial intelligence has greatly improved target identification, molecular modeling, and predictive analytics, but issues including data variability, interpretability, and regulatory obstacles continue to be unaddressed. AI platforms and tools show great promise in cutting time and expenses, yet their adoption into current workflows needs more testing and uniform guidelines. The review also highlights novel prospects, such as the merging of AI with multi-omics data and the increasing focus on interpretable AI for clinical application. We find AI-driven drug discovery has great potential, yet achieving its complete benefits requires collaboration across disciplines, strong validation systems, and flexible regulatory approaches. This study serves as a primary resource for scholars and professionals seeking to understand the dynamic field of artificial intelligence in pharmaceutical research.