Researchers produced 16 functional pathogens as experts warn the same technology could eventually pose serious biosecurity risks
US scientists have used artificial intelligence (AI) to create functional viruses that do not exist in nature for the first time, demonstrating a capability that researchers say could advance medicine but also pose serious biosecurity risks.
Researchers at Stanford University and the Broad Institute of MIT and Harvard used AI to design thousands of complete viral genomes, according to a study published in Science on Thursday.
Nearly 300 of the designs were chemically synthesized and tested in laboratories, producing 16 viable viruses.
The viruses were bacteriophages – organisms that infect bacteria rather than humans – and were based on a naturally occurring phage known as ΦX174. Some of the AI-designed versions proved more effective than natural phages at killing strains of E. coli that had developed resistance.
AI had previously been used extensively to design individual genes and proteins, but the researchers said this was the first demonstration that generative models could design an entire functional viral genome.
Fatemeh Vafaee, a biotechnology professor at the University of New South Wales, stressed that the phages posed no danger to humans but said the capability raised future security concerns.
“It’s less ‘should we worry about this virus’ and more ‘AI can now do this at all’,” she told Al Jazeera, calling for stronger oversight.
Experts noted, however, that designing dangerous human pathogens remains far more difficult than creating relatively simple bacteriophages.
The breakthrough comes amid growing concern over autonomous AI systems acting outside their intended instructions. Models developed by OpenAI, Anthropic, and Meta have recently carried out unauthorized cyber activity during security testing.
Britain’s AI Security Institute also reported this week that AI agents had targeted real people and organizations, created fake identities, and attempted to introduce malicious code during evaluations.
None of those incidents involved biological research, but they have raised concerns about increasingly capable AI systems being given access to powerful tools, including those used to design viruses.
The experiment also comes amid renewed debate in the US over dangerous virus research and the origins of Covid-19.
The possibility that the pandemic began with a laboratory accident at the US-funded Wuhan Institute of Virology has once again become a major political issue in the US following the recent congressional grilling of former White House coronavirus adviser Anthony Fauci.
Fauci was held in contempt by a US Senate committee on Thursday after refusing to answer questions about the origins of Covid-19 and US funding for coronavirus research in Wuhan.
US-funded biological research overseas has also come under scrutiny after former Director of National Intelligence Tulsi Gabbard released documents showing that Washington had supported biological laboratories in Ukraine researching dangerous pathogens, with some facilities involved in gain-of-function research.
Russia has long accused the US of setting up a network of biolabs in former Soviet states through Pentagon-linked programs. Moscow has repeatedly pointed to US-funded facilities in Ukraine, where research involved dangerous pathogens including anthrax, plague, and avian flu.
Read the full article here