ISSN print edition: 0366-6352 ISSN electronic edition: 1336-9075 Registr. No.: MK SR 9/7 Published monthly

Artificial intelligence for next-generation anxiolytic drug discovery: current trends, challenges, and future perspectives

Aliasgar Baldiwala, Azim M Baldiwala, Manan Patel, Pritkumar Amrutiya, Gopal Natesan, and Udit Chaube

Institute of Pharmacy, Nirma University, Ahmedabad, India

E-mail: uditchoube@gmail.com

Abstract:

Anxiety disorders are among the most prevalent psychiatric conditions, significantly affecting individuals’ quality of life and daily functioning. Despite the availability of various therapeutics, many patients continue to experience unresolved symptoms, underscoring the need for novel anxiolytic agents. This review highlights current trends and future perspectives in AI-driven anxiolytic drug discovery. It begins by discussing the limitations of traditional drug development approaches and outlines how Artificial intelligence (AI) techniques—including machine learning, deep learning, natural language processing, and generative models—are revolutionizing key stages such as target identification, virtual screening, pharmacokinetic profiling, and toxicity prediction. Emerging concepts like multi-modal data integration, explainable AI, and federated learning are proposed to enhance model transparency, collaboration, and trust. AI has become a disruptive technology in anxiolytic drug discovery, with AlphaFold being the first example, with its accurate predictions of CNS protein structure leading to a better target identification, RoseTTAFold and DiffDock being examples of tools used to model complex structure and ligand docking respectively. The review also addresses major challenges including data quality, validation protocols, and ethical considerations that must be resolved to ensure safe and effective AI applications. Furthermore, it explores the potential of innovations such as quantum computing, AI-human collaborative frameworks, and the incorporation of real-world and patient-reported outcomes. Ultimately, the future of anxiolytic treatments will be faster, more accurate, personalized, and cheaper with the help of AI applications.

Full paper is available at www.springerlink.com.

DOI: 10.1007/s11696-025-04598-0

Chemical Papers 80 (4) 3335–3358 (2026)