Researchers at Stanford University have developed an AI system capable of designing entirely new antibiotic compounds that are effective against drug-resistant bacteria. The system generated 12 candidate molecules, three of which showed potent activity against MRSA and other dangerous superbugs in laboratory testing.
The AI works by learning the molecular properties that give existing antibiotics their effectiveness and then exploring vast chemical spaces to design new structures with enhanced properties. The entire process, from design to initial lab validation, took just six weeks.
Antimicrobial resistance is projected to cause 10 million deaths annually by 2050 if new treatments are not developed. The Stanford team has partnered with a pharmaceutical company to begin preclinical development of the most promising candidates.
