|
Getting your Trinity Audio player ready...
|
Researchers at the Massachusetts Institute of Technology (MIT) have unveiled two new potential antibiotics, designed entirely by artificial intelligence, that show promise against drug-resistant gonorrhoea and MRSA. The discovery, published this week in the journal Cell, could signal the beginning of a “second golden age” in antibiotic development.
The two compounds, created atom-by-atom by generative AI, successfully killed the superbugs in laboratory experiments and animal tests. While years of refinement and clinical trials are still needed before these drugs reach patients, experts say the development marks a turning point in the decades-long struggle against antibiotic resistance.
Why It Matters
Drug-resistant infections are among the most urgent threats to global health, already causing more than a million deaths annually. Overuse and misuse of antibiotics have fueled the rise of superbugs that evade treatment, while innovation in new antibiotics has slowed dramatically.
“This is a critical advance,” said Prof. James Collins, a senior MIT researcher involved in the project. “We’re excited because we show that generative AI can be used to design completely new antibiotics. AI can enable us to come up with molecules cheaply and quickly, expanding our arsenal in the battle against superbugs.”
How AI Designed the Drugs
Unlike earlier approaches, which used AI to scan existing drug libraries, the MIT team pushed the boundaries by training AI models to invent new compounds from scratch. Researchers fed the system chemical structures of known antibiotics and data on how bacteria responded to them. The AI then generated millions of new molecular designs, filtering out toxic or ineffective candidates.
Two different strategies were tested: one that built molecules from smaller chemical fragments, and another that gave AI full creative freedom. Of the top theoretical designs, only a handful could be manufactured, but two compounds emerged as effective against gonorrhoea and methicillin-resistant Staphylococcus aureus (MRSA), both notorious for their resistance to existing treatments.
Expert Reactions
The scientific community has hailed the findings as groundbreaking but also cautioned against over-optimism.
“This is very significant with enormous potential,” said Dr. Andrew Edwards of Imperial College London’s Fleming Initiative. “It demonstrates a novel approach to identifying new antibiotics. However, while AI can speed up discovery, we still need the hard, time-consuming work of testing safety and efficacy.”
Prof. Chris Dowson of the University of Warwick called the study “cool” and a “significant step forward,” but warned of economic obstacles. “The irony is that the best antibiotics should be used sparingly to preserve their effectiveness, which makes them commercially unattractive. That creates a major challenge for getting new drugs to market.”
Next Steps in AI-Driven Drug Discovery
The two AI-designed antibiotics are not yet ready for human use. Researchers estimate another one to two years of laboratory refinement before clinical trials can even begin, a long, costly process with no guarantee of success. Manufacturing the drugs also poses challenges, as many AI-generated molecules prove difficult to synthesize at scale.
Still, the MIT team believes their work could reshape the future of medicine. “We’ve opened the door to an entirely new way of designing antibiotics,” Collins said. “With better models, AI could become a vital partner in addressing the global health crisis of antimicrobial resistance.”
As antibiotic resistance continues to rise, experts agree that bold innovations are needed. The promise of AI-driven drug discovery offers a rare glimpse of hope in a field where progress has long stalled, and could one day save millions of lives.
