Scientists at Stanford University have successfully employed artificial intelligence to design and create fully functional novel viruses in laboratory settings, representing the first instance of complete viral genomes being generated through AI technology. The research team used two AI models called Evo1 and Evo2, which operate similarly to large language models such as ChatGPT but predict genetic sequences rather than text. These models were trained on genetic codes from viruses, bacteria, plants, and people before being refined to generate bacteriophages—viruses that infect only specific bacterial species.
The researchers synthesized 302 AI-designed candidates and identified 16 that proved effective at killing E. coli bacteria in laboratory conditions. The resulting viruses pose no threat to humans, as they were engineered to infect bacteria exclusively. This breakthrough demonstrates a significant advancement in synthetic biology, the field concerned with designing organisms that go beyond those found in nature.
Scientists and experts have characterized the achievement as a major milestone with substantial implications for medicine and disease treatment. Potential applications include developing new therapies to combat antibiotic-resistant infections and creating novel drugs and treatments for various diseases. However, the technology has simultaneously sparked serious concerns among biosecurity experts regarding its potential misuse.
In commentary published alongside the research in Science journal, experts from Johns Hopkins University’s Center for Health Security raised what they described as urgent biosafety and biosecurity questions. They emphasized that the key issue is no longer whether AI can generate viral genomes, but rather how to prevent its use for causing harm. The Stanford researchers implemented multiple safety measures, including excluding viruses capable of infecting complex organisms from their training data and conducting all work in secure laboratory facilities.
Researchers noted that significant additional development would be required to apply this technology to more complex living organisms, as the bacteriophage genomes measure approximately 5,400 base pairs compared to the smallest living cell genomes at around 500,000 base pairs. Scientific experts from institutions worldwide have described the work as a transformative development that could lead to revolutionary approaches for addressing major health challenges.
Article Attribution | Read More at Article Source
Article summary produced by Claude AI