Scientists at the Massachusetts Institute of Technology (MIT) have used artificial intelligence (AI) to design potential new drugs against dangerous superbugs. These infections resist existing antibiotics, creating a growing global health crisis known as antimicrobial resistance (AMR).
Antimicrobial resistance occurs when bacteria, viruses, fungi, or parasites evolve so that medicines no longer work, making infections harder to treat. According to researchers, drug-resistant infections contributed to 4.71 million deaths in 2021, and this number is expected to rise.
The MIT team aimed to go beyond traditional approaches, using AI not just to speed up drug discovery but to create completely new molecules. Researchers generated more than 36 million potential compounds using generative algorithms before selecting the most promising candidates.
Their main targets were drug-resistant gonorrhoea, called an “urgent public health threat” by US officials, and multi-drug resistant Staphylococcus aureus (MDRSA), which includes the well-known MRSA.
One new compound, named NG1, proved highly effective against gonorrhoea bacteria in both laboratory dishes and mouse models. Meanwhile, six other molecules showed strong potential against MDRSA, suggesting fresh possibilities for tackling stubborn infections.
Researcher Aarti Krishnan explained the goal was to find new mechanisms, avoiding anything resembling existing antibiotics, in order to address resistance differently.
Professor James Collins, also from MIT, said the breakthrough demonstrates AI’s ability to explore vast chemical spaces never accessed before.
The findings, published in Cell, mark an important step in antibiotic development. The team is now working with nonprofit biotech Phare Bio to test these compounds further. If successful, they could move towards clinical trials, offering hope against the rising threat of superbugs.
