**
Researchers in the United States have achieved a remarkable milestone by employing artificial intelligence (AI) to design entirely new viruses capable of replication in laboratory settings. This groundbreaking achievement marks the first instance of AI successfully crafting complete viral genomes, presenting both exciting possibilities for medical advancements and urgent questions regarding biosecurity.
A New Frontier in Viral Design
The development involved the creation of 16 novel viruses specifically targeting bacterial infections, with no risk to human health. This advancement has been heralded as a “very significant turning point” in scientific research, potentially ushering in a new era of therapeutic strategies for combating diseases. However, experts have raised alarms about the implications of AI-engineered viruses, emphasizing the need for stringent safety measures.
Brian Hie, an assistant professor at Stanford University, remarked on the complexity of this achievement. “This is a next step in the complexity that’s designable by generative AI. It’s the first time generative AI has been used to design a complete genome that can replicate and perform functions within cells,” he stated. This innovation pushes the boundaries of synthetic biology, a field that intersects technology and life sciences.
Methodology Behind the Breakthrough
The AI models employed in this research, named Evo1 and Evo2, operate similarly to large language models like ChatGPT, but instead of predicting text, they forecast biological sequences. Trained on a diverse array of genetic information from viruses, bacteria, plants, and humans, these models were fine-tuned to create a type of virus known as bacteriophages, which are viruses that specifically infect bacteria.
From an initial pool of 302 designs generated by the AI, researchers synthesised 16 successful candidates in the lab, with a particular focus on their effectiveness in targeting E. coli bacteria. Samuel King, a PhD student in the laboratory, recounted the moment they realised the phages were functioning: “We were starting to see these clear spots, and it was just extremely exciting.” The enthusiasm in the lab was palpable, culminating in spontaneous applause when the results were shared with the broader team.
Implications for Antibiotic Resistance
The emergence of these new bacteriophages could provide innovative approaches to treating infections that have developed resistance to conventional antibiotics. As antibiotic resistance continues to escalate globally, alternative therapies such as phage therapy may offer viable solutions. This technology not only represents a potential breakthrough in the fight against stubborn bacterial infections but also highlights the transformative power of synthetic biology in healthcare.
However, the excitement surrounding these developments is tempered by a growing concern over the misuse of such technology. In a commentary accompanying the research published in the journal *Science*, Dr. Thomas Inglesby and Dr. Moritz Hanke from the Center for Health Security at Johns Hopkins University cautioned that the ability to create new viral genomes poses significant biosafety and biosecurity risks. They stressed the importance of careful oversight, asserting that the focus should remain on preventing the creation of viruses capable of causing harm.
The Path Forward
While the current research focused on non-pathogenic viruses, the potential exists for AI to evolve beyond merely designing bacteriophages. Hie acknowledges that while crafting living organisms remains a complex challenge, it is not beyond the realm of possibility. The smallest genome of a living cell comprises approximately 500,000 base pairs, compared to the phage genome of around 5,400 base pairs. The implications of successfully designing living entities could be profound, opening new avenues for treatment and genetic engineering.
Marc Güell, a professor in synthetic biology at Pompeu Fabra University, described the study as a “very significant turning point,” enabling researchers to “dream of exciting possibilities for tackling humanity’s greatest challenges.” He highlighted potential applications ranging from developing phages to combat diseases to engineering enzymes for genetic disorders and creating antibodies for immunotherapy.
Patrick Cai, chair of synthetic genomics at the Manchester Institute of Biotechnology, echoed this sentiment, noting that the findings suggest a nascent capability for AI to learn from evolutionary design principles. This insight may pave the way for AI-assisted genome writing, significantly advancing our understanding of biology.
Why it Matters
The implications of AI-designed viruses extend far beyond the laboratory. As we stand on the cusp of a new era in synthetic biology, the potential to innovate treatments for antibiotic-resistant infections could revolutionise public health. However, with great power comes great responsibility; the dual-use nature of this technology necessitates a robust framework for ethical governance and safety protocols. As we embrace these advancements, it is crucial to balance innovation with caution to ensure that the benefits to human health outweigh the risks of misuse.
