In a promising step towards tackling the global health crisis of antimicrobial resistance (AMR), researchers from the Indraprastha Institute of Information Technology Delhi (IIIT-Delhi), India, and Inria Saclay, France, have developed a powerful artificial intelligence (AI) tool. This new system is designed to help clinicians identify effective treatments for drug-resistant bacterial infections by repurposing existing antibiotics, potentially transforming infection management worldwide.Understanding the Threat of AMRAntimicrobial resistance occurs when bacteria evolve to resist the effects of antibiotics, making once-effective drugs useless. This growing issue threatens to turn minor infections like urinary tract infections or pneumonia into life-threatening conditions. The problem is especially severe in low- and middle-income countries, where over 70% of hospital-acquired infections are resistant to at least one widely used antibiotic.Adding to the challenge is the slow pace of antibiotic development. Creating a new drug often takes over ten years and involves substantial financial investment. As a result, scientists and clinicians are increasingly looking to drug repurposing—finding new uses for existing medications—as a faster, more cost-effective solution.How the AI Tool WorksTo support this strategy, a team led by Dr. Emilie Chouzenoux from Inria Saclay and Dr. Angshul Majumdar from IIIT-Delhi has developed a hybrid machine learning algorithm that can recommend treatment alternatives for resistant infections. Other key team members include research engineer Stuti Jain and graduate students Kriti Kumar and Sayantika Chatterjee.What sets this tool apart is its unique blend of clinical and molecular data. Instead of relying solely on rigid databases or predefined rules, the AI model learns from real-world treatment guidelines provided by top Indian hospitals. It then integrates this information with bacterial genomic data and the chemical structures of antibiotics to identify effective, lesser-known alternatives.Tested Against Deadly PathogensThe AI system was tested on several challenging pathogens known for their resistance to treatment:Klebsiella pneumoniae, which causes hospital-acquired pneumonia and bloodstream infectionsNeisseria gonorrhoeae, responsible for gonorrhea, which is increasingly difficult to treatMycobacterium tuberculosis, the bacterium that causes tuberculosisIn each case, the AI tool successfully recommended antibiotics that had either proven effectiveness or strong potential for repurposing. These results were validated by resistance data and clinical experts.A Global Solution for a Global Problem“This is an excellent example of how AI and international collaboration can come together to solve real-world medical challenges,” said Dr. Majumdar. He added that this approach makes better use of existing medical knowledge and enables quicker, smarter responses to the AMR crisis.Designed for scalability, the AI system can be integrated into hospital networks or public health programs, especially in settings with limited diagnostic tools. It not only aids in timely treatment decisions but also supports responsible antibiotic use.By linking clinical expertise with molecular science, this innovation marks a significant leap in the global fight against AMR—showcasing how collaborative, data-driven technologies can pave the way for better healthcare solutions.