Recent Approaches of Artificial Intelligence in Drug Discovery

Artificial intelligence (AI) and machine learning (ML) are bringing major changes to drug discovery
by helping to overcome long-standing problems such as high costs, slow development, and low
success rates. This review discusses recent progress (2019–2024) in the use of AI/ML throughout
the drug discovery process—from identifying drug targets to clinical development. It highlights
different techniques, including deep learning, graph neural networks, and transformers, and explains
how they are applied in areas such as target identification, lead discovery, hit optimization, and
safety testing. We compare the strengths and drawbacks of these approaches and outline key factors
needed for success, including high-quality data, reliable model validation, and ethical practices. The
review also points out current challenges, such as limited data access, poor model interpretability,
and difficulties in translating findings into clinical use. Finally, we suggest future directions to fully
realize the potential of AI in creating safer, more effective, and affordable medicines. The overall
aim is to promote responsible and transparent integration of AI into pharmaceutical research.