AI designs new superbug-killing antibiotics for gonorrhoea and MRSA

Authors: James Gallagher

Publisher:

BBC

Published: 8/14/2025

Language:English

--:--

Tom Banks

Good morning 跑了松鼠好嘛, I'm Tom Banks, and this is Goose Pod for you. Today is Saturday, August 16th.

Mask

I'm Mask. Today, we’re discussing a breakthrough: AI designing new superbug-killing antibiotics for gonorrhoea and MRSA.

Tom Banks

Let's get started. It’s a remarkable story out of MIT, where researchers used generative AI to design, atom-by-atom, new antibiotics. These aren't just discoveries; they are inventions aimed squarely at some of our most feared superbugs.

Mask

It's a fundamental shift in strategy. Forget searching for a needle in a haystack. The AI generated 36 million potential compounds, many of which have never existed. This is about creating the future of medicine, not just searching the past for answers. It's visionary.

Tom Banks

And it’s already showing promise. In lab and animal tests, two of these novel compounds successfully killed the superbugs. As Professor James Collins from the research team said, this gives us a much-needed 'leg up in the battle of our wits against the genes of superbugs.'

Mask

A 'leg up' is an understatement. This is the paradigm shift visionaries like Demis Hassabis have been working towards. We are accelerating evolution, moving from a slow, reactive process to one of proactive, intelligent design. This is how we win the arms race.

Tom Banks

You know, to really grasp the importance of this, we have to remember what it was like before antibiotics. A simple cut could be fatal. Then, after World War II, we entered a 'golden age' with penicillin and others. It truly felt like medicine had conquered infectious diseases.

Mask

But it was a false victory. We got complacent. For every miracle drug we introduced, bacteria began developing resistance almost immediately. Nature is the most relentless innovator, and it doesn't stop to celebrate. The microbes were already plotting their comeback.

Tom Banks

That’s right. Methicillin was created to fight penicillin-resistant staph, but within just three years, MRSA emerged. After that initial boom, the discovery of new antibiotic classes slowed to a crawl. The easy discoveries were made, and the pipeline started to run dry.

Mask

The pipeline didn't just run dry, it was abandoned! Pharmaceutical companies retreated because the economic model is fundamentally broken. It’s incredibly expensive and slow to develop a drug that, if successful, you’ll tell doctors to use as sparingly as possible. It’s a terrible business proposition.

Tom Banks

And that created this dangerous void we're in now, with bacteria evolving faster than our ability to find new ways to fight them. It’s this very crisis that makes the new AI approach so incredibly vital and timely. It’s a potential solution to a problem decades in the making.

Mask

Exactly. The central conflict has been the commercial viability. Traditional methods are too slow and costly. AI attacks that problem head-on by accelerating discovery and dramatically lowering the initial R&D costs. It’s not just a new tool; it's a new economic engine for antibiotic development.

Tom Banks

But there's an ethical tension here, too. When we talk about AI designing molecules or even resurrecting ancient genes for antibiotics, it raises profound questions. Are there unforeseen ecological risks? We have to balance the rush for solutions with caution and responsibility.

Mask

Caution is one thing, paralysis is another. Every great leap forward comes with risks. The far greater risk is doing nothing while superbugs kill millions. We need strong biosafety protocols, not a failure of nerve. The potential to save humanity outweighs the hypothetical fears.

Tom Banks

And there are practical hurdles. The researchers noted that even though the AI designed 80 promising molecules for gonorrhoea, they only had the resources to actually manufacture and test two of them. The brilliance of the AI is still limited by real-world chemistry and costs.

Tom Banks

Assuming we navigate those challenges, the public health impact could be staggering. We're talking about blunting a crisis that causes over a million deaths each year. For families affected by resistant infections, this research represents a profound sense of hope. It’s truly about saving lives.

Mask

And the economic impact is just as revolutionary. AI-designed drugs are showing 80-90% success rates in early trials, compared to 40-65% for traditional drugs. This efficiency doesn't just produce one new medicine; it has the potential to fix the entire broken, unprofitable pipeline.

Tom Banks

That improved success rate is key. It means more effective and safer medicines can get to the people who need them much faster. For a patient, that’s not just a statistic, it’s a second chance at life they might not have had otherwise.

Mask

This is only the beginning. AI has already found other novel compounds, like Halicin and thousands of molecules called 'archaeasins,' that seem to work in entirely new ways. We are on the verge of unlocking a vast, undiscovered pharmacy that exists only within data.

Tom Banks

So the future isn't just about finding a single new drug, but about creating a sustainable engine for discovery. It’s the dawn of that 'second golden age' of antibiotics the MIT team talked about, keeping us one step ahead in this critical race.

Tom Banks

That’s the end of today’s discussion. The main takeaway is AI’s incredible potential to revolutionize our fight against superbugs.

Mask

Thank you for listening to Goose Pod. See you tomorrow.

## AI Designs Novel Antibiotics to Combat Drug-Resistant Superbugs This news report from the **BBC**, authored by **James Gallagher**, details a groundbreaking advancement in antibiotic discovery, where artificial intelligence (AI) has successfully designed two new potential antibiotic compounds. These compounds have demonstrated the ability to kill drug-resistant strains of **gonorrhoea** and **MRSA (methicillin-resistant Staphylococcus aureus)** in laboratory and animal tests. ### Key Findings and Conclusions: * **AI-Designed Antibiotics:** Researchers at the **Massachusetts Institute of Technology (MIT)** have utilized generative AI to design entirely new antibiotic molecules, atom-by-atom. This marks a significant step beyond previous AI applications that focused on identifying existing chemicals with antibiotic potential. * **Effectiveness Against Superbugs:** The two AI-designed compounds have shown efficacy in killing drug-resistant gonorrhoea and MRSA in laboratory settings and in infected mice. * **Potential for a "Second Golden Age":** The MIT team believes AI could usher in a new era of antibiotic discovery, addressing the critical shortage of new drugs to combat rising antibiotic resistance. * **Addressing a Global Health Crisis:** Antibiotic-resistant infections are a growing concern, causing over a million deaths annually. The overuse of antibiotics has accelerated bacterial evolution, making existing treatments less effective. ### Key Statistics and Metrics: * **Interrogated Compounds:** The AI was trained on and interrogated **36 million compounds**, including those that do not yet exist. * **Compound Size:** The AI identified promising starting points by searching through a library of chemical fragments ranging from **eight to 19 atoms** in size. * **Manufacturing Challenges:** Out of the top 80 theoretical gonorrhoea treatments designed by AI, only **two** were successfully synthesized into actual medicines, highlighting manufacturing challenges. ### Important Recommendations and Future Steps: * **Further Refinement and Clinical Trials:** The newly designed compounds are not yet ready for prescription. They require an estimated **one to two years** of further refinement before they can enter clinical trials in humans. * **Improved AI Models:** There is a need for better AI models that can more accurately predict drug effectiveness within the human body, moving beyond laboratory performance. ### Significant Trends and Changes: * **Shift in AI Application:** The research signifies a shift from AI being used to screen existing chemicals to AI being used for the *de novo* design of novel drug molecules. * **Accelerated Discovery Process:** AI has the potential to significantly speed up the drug discovery process, enabling the creation of new molecules "cheaply and quickly." ### Notable Risks and Concerns: * **Long and Expensive Testing:** Despite AI's capabilities, the process of testing for safety and efficacy in humans remains long, expensive, and without a guarantee of success. * **Manufacturing Feasibility:** The complexity of AI-designed molecules can pose challenges in their synthesis and manufacturing. * **Economic Viability:** A significant economic concern is the profitability of new antibiotics. To preserve their effectiveness, these drugs should ideally be used sparingly, making it difficult for pharmaceutical companies to recoup development costs. ### Context and Expert Opinions: * **Prof James Collins (MIT):** Emphasizes AI's ability to generate novel molecules quickly and cheaply, bolstering the fight against superbugs. * **Dr Andrew Edwards (Fleming Initiative and Imperial College London):** Praises the work as "very significant" with "enormous potential" but stresses the continued need for rigorous safety and efficacy testing. * **Prof Chris Dowson (University of Warwick):** Describes the study as "cool" and a "significant step forward," but also points to the economic disincentive for developing new antibiotics. This research represents a significant leap forward in the battle against antibiotic resistance, showcasing the transformative potential of AI in drug discovery. However, the path from AI design to patient prescription remains a complex and challenging one, requiring substantial further research and development.

Read original at BBC →

Getty ImagesArtificial intelligence has invented two new potential antibiotics that could kill drug-resistant gonorrhoea and MRSA, researchers have revealed.The drugs were designed atom-by-atom by the AI and killed the superbugs in laboratory and animal tests.The two compounds still need years of refinement and clinical trials before they could be prescribed.

But the Massachusetts Institute of Technology (MIT) team behind it say AI could start a "second golden age" in antibiotic discovery.Antibiotics kill bacteria, but infections that resist treatment are now causing more than a million deaths a year.Overusing antibiotics has helped bacteria evolve to dodge the drugs' effects, and there has been a shortage of new antibiotics for decades.

Researchers have previously used AI to trawl through thousands of known chemicals in an attempt to identify ones with potential to become new antibiotics.Now, the MIT team have gone one step further by using generative AI to design antibiotics in the first place for the sexually transmitted infection gonorrhoea and for potentially-deadly MRSA (methicillin-resistant Staphylococcus aureus).

Their study, published in the journal Cell, interrogated 36 million compounds including those that either do not exist or have not yet been discovered.Scientists trained the AI by giving it the chemical structure of known compounds alongside data on whether they slow the growth of different species of bacteria.

The AI then learns how bacteria are affected by different molecular structures, built of atoms such as carbon, oxygen, hydrogen and nitrogen.Two approaches were then tried to design new antibiotics with AI. The first identified a promising starting point by searching through a library of millions of chemical fragments, eight to 19 atoms in size, and built from there.

The second gave the AI free rein from the start.The design process also weeded out anything that looked too similar to current antibiotics. It also tried to ensure they were inventing medicines rather than soap and to filter out anything predicted to be toxic to humans.Scientists used AI to create antibiotics for gonorrhoea and MRSA, a type of bacteria that lives harmlessly on the skin but can cause a serious infection if it enters the body.

Once manufactured, the leading designs were tested on bacteria in the lab and on infected mice, resulting in two new potential drugs.MITProf James Collins, one of the researchers at MIT"We're excited because we show that generative AI can be used to design completely new antibiotics," Prof James Collins, from MIT, tells the BBC."

AI can enable us to come up with molecules, cheaply and quickly and in this way, expand our arsenal, and really give us a leg up in the battle of our wits against the genes of superbugs."However, they are not ready for clinical trials and the drugs will require refinement – estimated to take another one to two year's work – before the long process of testing them in people could begin.

Dr Andrew Edwards, from the Fleming Initiative and Imperial College London, said the work was "very significant" with "enormous potential" because it "demonstrates a novel approach to identifying new antibiotics".But he added: "While AI promises to dramatically improve drug discovery and development, we still need to do the hard yards when it comes to testing safety and efficacy."

That can be a long and expensive process with no guarantee that the experimental medicines will be prescribed to patients at the end.Some are calling for AI drug discovery more broadly to improve. Prof Collins says "we need better models" that move beyond how well the drugs perform in the laboratory to ones that are a better predictor of their effectiveness in the body.

There is also an issue with how challenging the AI-designs are to manufacture. Of the top 80 gonorrhoea treatments designed in theory, only two were synthesised to create medicines.Prof Chris Dowson, at the University of Warwick, said the study was "cool" and showed AI was a "significant step forward as a tool for antibiotic discovery to mitigate against the emergence of resistance".

However, he explains, there is also an economic problem factoring into drug-resistant infections - "how do you make drugs that have no commercial value?"If a new antibiotic was invented, then ideally you would use it as little as possible to preserve its effectiveness, making it hard for anyone to turn a profit.