Researchers have developed antibiotics that can potentially kill drug-resistant gonorrhoea and MRSA with the help of artificial intelligence.
The team at the Massachusetts Institute of Technology (MIT) used generative AI algorithms to design more than 36 million molecular compounds, identifying two new drugs that were then tested in lab settings and in mice.
According to their study, published in Cell on Thursday, the AI scanned through a list of potential molecules and predicted which ones could work as antibiotics, while also avoiding structures that could be harmful to humans or too similar to existing drugs.
Both new antibiotics were able to kill strains of bacteria that are becoming increasingly difficult to treat with existing antibiotics.
One of the new drugs, called NG1, was designed to specifically target gonorrhoea, a sexually transmitted infection that has been rapidly developing resistance to antibiotics. The World Health Organisation has warned that gonorrhoea is becoming untreatable in some cases due to antibiotic resistance.
New data from the UK Health Security Agency (UKHSA) reveals a concerning increase in resistant cases. Between January 2024 and March 2025, 17 cases of ceftriaxone-resistant gonorrhoea were reported, 13 in 2024 and four in the first three months of 2025.
The second drug, named DN1, was effective against methicillin-resistant staphylococcus aureus (MRSA), a type of staph infection that no longer responds to many commonly used antibiotics.
MRSA often begins as a painful, pus‑filled skin boil and can spread to cause serious, hard‑to‑treat infections in the bloodstream, lungs or bones, especially in people with weakened immune systems.
After tests, it was found that NG1 and DN1 stood out for their ability to kill dangerous bacteria in both test tubes and mice. In animal studies, the drugs were able to successfully treat skin infections caused by MRSA and clear gonorrhoea infections.
The team also hopes to use their AI platform to design new drugs for other superbugs in the future, including those that cause tuberculosis and hospital-acquired infections like Pseudomonas aeruginosa.
“We’re excited about the new possibilities that this project opens up for antibiotics development,” said James Collins, senior author of the study and Termeer Professor of Medical Engineering and Science in MIT’s Institute for Medical Engineering and Science (IMES) and Department of Biological Engineering.
“Our work shows the power of AI from a drug design standpoint, and enables us to exploit much larger chemical spaces that were previously inaccessible.”
What makes this discovery significant is that both drugs work in new ways. Many antibiotics rely on similar mechanisms to kill bacteria, which has allowed some bacteria to evolve resistance over time. These new compounds appear to attack bacterial membranes in ways not used by current medicines, giving scientists hope that they could be harder for bacteria to resist.
“We wanted to get rid of anything that would look like an existing antibiotic, to help address the antimicrobial resistance crisis in a fundamentally different way,” said Aarti Krishnan, lead author and MIT postdoc.
Although the findings are promising, the researchers have stressed that these new antibiotics are still in early development. They must go through years of testing to prove they are safe and effective in humans.
The next steps include refining the chemical structure of the drugs, testing for potential side effects, and eventually conducting clinical trials.
Antimicrobial resistance, or when bacteria evolve to resist the drugs used to kill them, is a growing public health crisis. It is estimated to be responsible for more than one million deaths globally each year, with experts warning that without urgent action, even minor infections could once again become deadly.
In the UK, there were 66,730 serious antibiotic-resistant infections recorded in 2023 alone, according to a report by The Times.
Why the students not getting into uni this year could be the lucky ones
How cancer could soon be detected using just a voice note
Musk’s Grok calls Trump the ‘most notorious’ criminal in DC when asked about crime
Hair strand stuck in ancient device could change what’s known about Inca civilisation
Scot, 12, named on Time’s first Girls of the Year list after homelessness design
Shark-like prehistoric whale with razor-sharp teeth discovered off Australia
