- Scientists have created the first viruses designed by artificial intelligence (AI), advancing phage therapy.
- The research raises important biosafety and biosecurity concerns, prompting calls for governance frameworks.
- Generative AI models were used to design functioning genomes of bacteriophages, with limited risk from dangerous pathogens during training.
- Experts stress the need for a multi-layered approach in regulating genetic engineering.
Breakthrough in Virus Design
Scientists have achieved a groundbreaking milestone by designing viruses using artificial intelligence (AI) for the first time. Researchers at Stanford University, led by chemical engineer Dr. Brian Hie, used AI models to generate genetic codes that were then manufactured into functional bacteriophages—viruses specifically designed to infect bacteria.
These engineered bacteriophages demonstrated efficacy in overcoming antibiotic-resistant E coli strains. A cocktail of the AI-designed phages effectively killed these resistant bugs, offering a potential new avenue for treating persistent infections with bacterial resistance.
Risk and Regulation
The accomplishment brings both promise and peril. The scientists emphasized that while the technology holds significant potential in transforming phage therapy and expanding biotechnological tools, it also raises urgent biosafety, biocontainment, and biosecurity considerations. They urged others working on similar projects to consult safety and security professionals.
In a supporting article, Prof. Tom Inglesby and Dr. Moritz Hanke from the Center for Health Security at Johns Hopkins University warned that the ability of AI to compose viral genomes now exists, necessitating robust governance frameworks to ensure safe application. They highlighted the potential for misuse, particularly in creating pathogens that might not be contained by current measures.
Expert Opinions
Dr. Filippa Lentzos from King's College London stressed the importance of intervention points such as DNA synthesis screening and responsible research review. She advocated for a layered approach to governance, including safeguards around model development and access, along with established laboratory biosafety and biosecurity measures.
Challenges Ahead
The work on AI-designed viruses is still in its early stages, with complexity increasing as genome sizes grow. However, experts like Prof. Tom Ellis at Imperial College London noted that while the risk of full AI design for harmful pathogens is overblown compared to existing gain-of-function changes, it remains a critical area requiring close regulation.
The researchers and experts agree on the need for robust governance frameworks to balance innovation with safety, ensuring that advances in genetic engineering do not come at the cost of public health risks. As AI capabilities continue to evolve, the challenge lies in developing regulatory mechanisms that can adapt quickly to new technologies.
Source: The Guardian
Science Daily
158 posts
