AI model Evo creates viable synthetic viruses in lab
Scientists at Stanford University and the Arc Institute trained an artificial‑intelligence system called Evo on DNA sequences from millions of organisms. The model learned the grammatical rules of genetic code and generated designs for entirely new viral genomes. Researchers synthesized 285 DNA molecules based on the AI's proposals, of which 16 produced viable bacteriophages that can infect bacteria but not humans. The viruses resemble the natural bacteriophage Phi X‑174. The findings were published in the journal Science on 6 August 2026.
The breakthrough demonstrates AI's potential to accelerate medical research, such as vaccine and antiviral development, while also sparking biosecurity concerns about the ease of creating novel pathogens that could be weaponised.
Entities: Arc Institute · Evo (AI model) · Patrick Cai · Phi X-174 bacteriophage · Stanford University
Claims
What the coverage asserts, and how well corroborated each claim is across sources.
- [● 3 SOURCES] An AI model named Evo was trained on DNA sequences from millions of organisms. (Study authors)
- [● 3 SOURCES] The AI designed 16 viable viral genomes that were synthesized in the laboratory. (Study authors)
- [● 3 SOURCES] The resulting viruses are bacteriophages that infect only bacteria and pose no threat to humans. (Study authors)
- [● 2 SOURCES] Researchers from Stanford University and the Arc Institute in Palo Alto, California conducted the work. (Study authors)
- [○ 1 SOURCE] The AI‑generated viruses are similar to the natural bacteriophage Phi X‑174. (Study authors)
- [● 3 SOURCES] The study was published in the journal Science on 6 August 2026. (Study authors)
- [○ 1 SOURCE] During training, the AI model analyzed approximately nine thousand billion nucleotides. (Study authors)
- [● 3 SOURCES] The research raises concerns about the potential misuse of AI for creating bioweapons. (Study authors)