Researchers at Stanford have successfully created synthetic viruses that can infect and kill bacteria using a generative AI model similar to ChatGPT. These viruses, known as bacteriophages, have the potential to combat antibiotic resistance and could lead to new treatments for bacterial infections.
The AI model, called Evo, was trained on a vast amount of genomic data and can generate coherent genome-scale sequences. In the study, Evo designed thousands of potential virus genomes, and 16 of them were able to replicate, infect, and destroy bacteria.
The synthetic viruses demonstrated versatility, with a combination of them able to infect and kill multiple strains of bacteria, even those they were not designed to target. This technology could lead to the creation of new treatments for bacterial infections and potentially revolutionize the field of synthetic biology.
However, the use of generative AI in synthetic biology also raises concerns about biosafety and the potential risks of creating new life forms. Further testing and development are needed to fully realize the potential of AI-designed viruses.
The study highlights the potential of AI in advancing synthetic biology and could lead to breakthroughs in the development of new treatments for bacterial infections. As researchers continue to explore the capabilities of AI in this field, it's likely that we'll see significant advancements in the coming years.
