AI Creates 16 Viable Viruses Never Seen in Nature

AI-designed viruses demonstrate synthetic biology breakthrough

Researchers have successfully used artificial intelligence to design and create functional viruses that do not exist in nature, according to a study published Thursday in the journal Science. Scientists at the Arc Institute in Palo Alto, California, trained an AI model to recognize DNA structural patterns and generate entirely new viral genome recipes.

The team synthesized DNA molecules based on the AI's designs and inserted them into bacteria. The modified bacteria then produced 16 viable viruses capable of infecting other bacteria, demonstrating that the AI-created pathogens were fully functional biological entities.

While scientists have long been able to manufacture copies of existing viral genomes for research into vaccines and antiviral drugs, this work represents the first time AI has been used to invent completely novel viruses from the ground up, as first reported by The New York Times.

Why it matters

This breakthrough opens new pathways for developing medical treatments and understanding viral mechanisms, but it also introduces biosecurity concerns about the potential misuse of AI to engineer dangerous pathogens. The technology demonstrates that machine learning can now design functional biological systems without human-directed engineering of specific traits—a capability that could accelerate both beneficial research and create new risks requiring governance frameworks.

Technical approach and results

The Arc Institute researchers trained their AI model on libraries of DNA sequences, teaching it to identify patterns in how natural viruses are structured. The model then generated recipes for viral genomes that had never existed in nature.

All 16 successfully created viruses were similar to Phi X-174, a naturally occurring virus that infects only bacteria and poses no threat to humans. This design choice limited the immediate risk while demonstrating the proof of concept.

Expert perspective

Patrick Cai, a synthetic biologist at the University of Manchester who was not involved in the research, called the work "an important milestone" in the field.

The study advances synthetic biology beyond simply duplicating existing genetic material. By using AI to generate novel designs, researchers can now explore viral architectures that evolution has not produced, potentially uncovering new approaches to fighting disease or developing biotechnology tools.

The findings were first reported by Carl Zimmer in The New York Times and published in the journal Science.