As antimicrobial resistance (AMR) continues to threaten global health, a collaborative team of researchers from France and India has introduced a cutting-edge artificial intelligence (AI)-based method to help tackle the problem. The innovative system, developed jointly by Inria Saclay in France and the Indraprastha Institute of Information Technology Delhi (IIIT-Delhi), is designed to recommend alternative antibiotics for treating drug-resistant bacterial infections, using the principle of repurposing existing drugs.AMR arises when bacteria evolve to withstand antibiotics that once killed them. This makes routine infections such as urinary tract infections, pneumonia, and wound infections more difficult—and sometimes impossible—to treat. According to recent data, more than 70% of hospital-acquired infections in low- and middle-income countries are resistant to at least one commonly used antibiotic, posing serious challenges to public health systems.The traditional route of developing new antibiotics is often slow, costly, and uncertain, taking over a decade and massive financial resources. In light of these limitations, the scientific community is increasingly turning to drug repositioning—finding new uses for existing medicines—as a more feasible solution.In this context, the AI-based system developed by the France-India research team could mark a breakthrough. Led by Dr. Emilie Chouzenoux from Inria Saclay and Dr. Angshul Majumdar from IIIT-Delhi, the project also involves research engineer Stuti Jain and graduate students Kriti Kumar and Sayantika Chatterjee.The team’s AI model uses a hybrid machine learning approach, moving beyond fixed rule-based systems. It draws on real-world clinical data, including antibiotic usage protocols sourced from top Indian hospitals, offering a nuanced view of current treatment practices. The system is further enriched by molecular data—such as bacterial genome sequences and chemical profiles of antibiotics—to pinpoint lesser-known or underutilised drugs that may prove effective against resistant strains.The AI method was tested in case studies involving multidrug-resistant bacteria like Klebsiella pneumoniae, Neisseria gonorrhoeae, and Mycobacterium tuberculosis—organisms known for causing hospital infections, sexually transmitted diseases, and tuberculosis. In each scenario, the AI tool suggested potentially effective antibiotics, which were then validated against existing resistance databases and expert evaluations.“This is an excellent example of how AI and international collaboration can come together to solve real-world medical challenges,” said Dr. Majumdar. “Our method enables smarter, faster decision-making using existing knowledge, and brings us closer to curbing the global AMR threat.”The researchers believe the AI system could be especially useful in hospitals and public health settings, helping doctors and microbiologists reduce treatment delays while promoting more responsible antibiotic use. It may also prove beneficial in regions with limited diagnostic capabilities, offering clinical support where resources are scarce.With further development, the team envisions the tool becoming part of routine infection management, enhancing antibiotic stewardship and contributing to the global fight against antimicrobial resistance.