DAILY SCIENCE NEWS
Health & Medicine ✓ Peer-Reviewed & Fact-Checked

Artificial Intelligence Used to Create a Virus for the First Time in Stanford University Breakthrough

Scientists at Stanford University have for the first time successfully designed and produced a new bacteriophage virus using AI models Evo1 and Evo2, marking a major medical breakthrough while raising urgent global biosafety concerns.

✍️ By: Daily Science News 📅 Published: August 10, 2026 ⏱️ Read Time: 7 min read
Artificial Intelligence Used to Create a Virus for the First Time in Stanford University Breakthrough - Science News

1.AI Creates a Virus for the First Time in History

Scientists have, for the first time in history, successfully designed and produced a new virus using artificial intelligence. While hailed as a landmark breakthrough for the future of drug discovery and personalized medicine, the achievement has simultaneously stirred deep concern within the global medical and biosafety community about the responsible use of this technology and its potential future risks if misused.

2.Dr. Brian Hie and the Stanford Research Team

The virus was created by a research team led by Dr. Brian Hie, a chemical engineer at Stanford University in California. The team engineered a type of virus known as a bacteriophage — a virus that infects only bacteria and is widely used around the world in phage therapy to treat patients suffering from chronic, antibiotic-resistant bacterial infections. In laboratory tests, the AI-generated viral formulation successfully destroyed strains of E. coli bacteria that had shown strong resistance to naturally occurring viruses.

3.How Evo1 and Evo2 AI Models Designed the Genome

To design the genome of this new virus, the researchers used AI models known as Evo1 and Evo2, trained on genetic data from more than two million bacteriophages. To minimize the risk of creating dangerous viruses, the researchers deliberately excluded genetic code from any viruses known to affect humans, animals, or plants during training. This careful exclusion was designed to create a safety boundary around the AI's generative capabilities.

4.From Thousands of Candidates to 16 Successful Viruses

Of the thousands of potential genomes generated by the AI models, researchers selected around 300 to actually synthesize in the laboratory — of which only 16 proved successful. This relatively low success rate underscores the complexity of viral biology and highlights that, while AI dramatically accelerates the design phase, the path from digital genome to functional living virus still involves substantial biological unpredictability.

5.Transformative Potential for Medicine and Biotechnology

According to the research report published in the journal Science, the ability to rapidly design viral genomes could bring about a major transformation in medicine and biotechnology. Bacteriophages are already used in phage therapy to treat drug-resistant bacterial infections — a growing global health problem as antibiotic resistance increases. AI-designed phages could be customized quickly and precisely for specific bacterial strains, opening an entirely new era of personalized infectious disease treatment.

6.Urgent Biosafety Warnings from Global Experts

However, the scientific community has not been slow to raise alarm. Professor Tom Inglesby of Johns Hopkins University and Dr. Moritz Heinke warn that the same technology could eventually be used to design far more dangerous pathogens. Professor Tom Ellis of Imperial College London called the discovery fascinating, but noted that designing more complex genomes remains a difficult task for AI at this stage. Dr. Filippa Lentzos of King's College London stressed the absolute necessity of strict international regulation and oversight in the production of potentially dangerous synthetic DNA. The world currently lacks established regulations or control mechanisms to govern the safe use of AI in virus design — a critical gap that this breakthrough has sharply exposed.

Editorial Standards & Fact-Checking Transparency

This report adheres to Daily Science News' rigorous academic guidelines. All claims are verified against primary scientific literature from institutions including NASA, ESA, CERN, and peer-reviewed journals. Supervised by Dulaksha Sandeepa. Questions or corrections? Contact contact@sciencenewshub.click.