Infections · Drug discovery
Generative AI designs new antibiotics against superbugs
MIT researchers used generative AI to design brand-new antibiotic molecules, published in Cell in August 2025. Two candidates worked against drug-resistant gonorrhoea and MRSA in the lab and in mice. It's early research: no human has taken them.
How far away is this?
Lab and animal results only. Most early drug candidates never reach patients.
Where is it approved?
| Region | Regulator | Status | Notes |
|---|---|---|---|
| United States | FDA | Not approved | Lab and mouse studies only. |
| European Union | CE mark / EMA | Not approved | Lab and mouse studies only. |
| United Kingdom | MHRA | Not approved | Lab and mouse studies only. |
| Australia | TGA | Not approved | Lab and mouse studies only. |
Lab and mouse studies only; no human has taken these compounds. "Unknown" means we haven't verified it, not that it's unavailable. Approval in one country isn't approval in another.
What is it?
Instead of screening existing chemicals, the AI invented new molecules predicted to kill bacteria, which the team then made and tested.
What's the evidence?
A compound called NG1 was active against multidrug-resistant Neisseria gonorrhoeae, and DN1 against MRSA, including in mouse infection models.[3]
Evidence stage: Early research · #c1
The study was published in Cell (Krishnan et al., 2025).[1]
Evidence stage: Early research · #c2
What does it mean for me?
Antibiotic resistance is a real worry, and new tools help. But these aren't medicines yet. Using antibiotics only when they're needed still matters.
What could I ask my doctor?
- Do I need an antibiotic for this?
- How can I lower my risk of a resistant infection?
These are general starting points, not a recommendation to choose any particular test or treatment.
Change history
- : First published.
- : Updated after independent fact-check and clinical review
- : Review label corrected: this page was not clinically reviewed. Earlier labels saying 'independent fact-check' and 'clinical review' were wrong; checks so far were AI-assisted source checks only.
General information only, not medical advice. Talk to your doctor about your situation. AI-assisted source check. Not clinically reviewed.
References
- Krishnan et al., Cell (2025). cell.com (2025). Peer-reviewed study. https://www.cell.com/cell/abstract/S0092-8674(25)00855-4
- PubMed 40816267. pubmed.ncbi.nlm.nih.gov. Abstract. https://pubmed.ncbi.nlm.nih.gov/40816267/
- MIT News. news.mit.edu. University news. https://news.mit.edu/2025/using-generative-ai-researchers-design-compounds-kill-drug-resistant-bacteria-0814
Publisher shown as the website hosting each source. Dates are given where the source label includes them.