Artificial Intelligence has recently demonstrated the ability to create functional viruses, designed for biological purposes rather than digital ones.
Stanford bioengineers, led by chemical engineering professor Brian Hie, have made a significant breakthrough in the field of synthetic biology. They have successfully created synthetic bacteriophages using AI-generated designs, a first in the real world.
According to Hie, designing a bioweapon with AI would be much harder than taking something from nature. This is because the work required to create AI-generated bacteriophages is extensive, involving both computational work and experimental work. This, Hie argues, does not lower the barrier of access to designing a bioweapon, as is the common litmus test in the biosecurity community.
The team's research was funded by the National Institutes of Health (NIH) and used the Evo 1 and Evo 2 models from the Arc Institute. Hie contributed to the design of both Evo models. The AI models were not trained on human virus genome data.
The synthetic bacteriophages were created using ΦX174, a bacteriophage that targets E. coli bacteria, as a starting point. Sixteen of the synthetic bacteriophages were found to inhibit the growth of E. coli bacteria. Among them, the synthetic bacteriophage Evo-Φ69 showed an expansion rate between 16 fold and 65 fold over a six-hour infection period, significantly outperforming ΦX174.
Several of the AI-generated phages developed in this project qualify as entirely new species according to phage taxonomy. The end result of the AI portion of the project was a pool of 302 candidate genomes, 285 of which were able to generate full genomes.
The team has posted a paper on the preprint service bioRxiv detailing their research. The paper is currently under peer review. However, concerns about the potential misuse of the AI-generated phages for creating deadly human viruses have been raised.
Biotechnology journalist Niko McCarty expressed his concerns, pointing out that the HIV genome is only about 10,000 bases in length, not much larger than bacteriophages, and the coronavirus genome is about 30,000 bases.
Despite these concerns, Hie is not worried about the potential misuse of the AI-generated phages for creating deadly human viruses. He argues that the work required by his team to get from Evo to artificial bacteriophages makes it harder for a bad actor to design a bioweapon using AI.
The team suggests that generative AI could be used to produce genetically diverse phage cocktails that could improve therapeutic efficacy, particularly for multidrug resistant bacterial infections. This could pave the way for new treatments and therapies in the future.
As the field of synthetic biology continues to evolve, the potential applications of AI-generated organisms are vast and exciting. However, it is crucial that the community continues to address and discuss the ethical implications and potential risks associated with this technology.
Read also:
- Enhanced Iron Absorption in Female Health: Biotechnology Developed Plant Protein Outperforms Iron Supplements in Fermentation
- Controversy Surrounding Epstein Heats Up in Washington; Trump Endorses Homelessness Executive Order; More Events Reported
- Prevent the exploitation of our public health care systems for financial gain
- Announced winners of the American Library Association Awards in 2025
