Researchers at Stanford University and the Arc Institute have successfully created new self-replicating viruses using genetic sequences designed by artificial intelligence, marking a significant development in artificial intelligence-assisted biological research. The study, published in Science, used genome language models called Evo 1 and Evo 2, which are trained to understand and generate genetic code. The researchers used the models to produce potential genomes for bacteriophages, viruses that infect bacteria. Laboratory testing confirmed that some of the artificial designs were viable and capable of replicating inside E. coli bacteria. The researchers said the significance of the work lies in demonstrating that artificial intelligence can generate complete genetic designs for biological entities that do not occur naturally. The technology could eventually support applications such as targeted phage therapy and the development of specialised biological tools. However, scientists have also raised biosafety and biosecurity concerns, warning that increasingly capable biological design systems could potentially be misused. The study highlights the growing need for responsible safeguards as artificial intelligence capabilities advance in life sciences
Read more here
