In a groundbreaking development, scientists have used artificial intelligence to design and create 16 new viruses, marking a significant milestone in the application of AI to biotechnology. The research, published in early 2026, demonstrates how AI can rapidly generate novel bacteriophages—viruses that infect bacteria—offering a potential new weapon in the battle against antibiotic-resistant infections.
The Promise of AI-Designed Phages
Bacteriophages, or phages, are viruses that specifically target and kill bacteria. With antibiotic resistance rising globally, phages have emerged as a promising alternative therapy. However, finding or engineering phages that are safe and effective has traditionally been a slow, labor-intensive process. AI changes this by enabling the design of custom phages in a fraction of the time.
Researchers at Profluent, a California-based biotech company, used a large language model trained on vast datasets of biological sequences to generate thousands of potential phage designs. From these, they selected and synthesized 16 that showed the most promise. The resulting viruses were validated in laboratory tests, demonstrating their ability to infect and kill specific strains of bacteria, including some resistant to conventional antibiotics.
How It Works: AI as a Biological Engineer
The AI system, similar in architecture to models used for text generation, learned the “grammar” of viral genomes. By analyzing millions of existing phage sequences, it identified patterns associated with infectivity and host specificity. When prompted to design new phages, it produced sequences that not only were novel but also functional—a feat that has eluded traditional computational methods.
“This is like having a master engineer who can draft blueprints for new viruses on demand,” said Dr. Emily Carter, a lead researcher on the project. “The implications for medicine are enormous, but so are the responsibilities.”
Regulatory and Ethical Challenges
While the potential therapeutic benefits are clear, the ability to create custom viruses raises serious biosecurity concerns. Experts worry that the same technology could be misused to design pathogens for harmful purposes, and that current regulatory frameworks are not equipped to handle the speed and scale of AI-driven biological design.
In 2026, the World Health Organization (WHO) and national agencies are scrambling to update guidelines for AI-assisted biotech research. The new U.S. Biosecurity Act, passed in late 2025, requires oversight of AI systems capable of generating dangerous biological agents, but critics argue it lacks teeth and global coordination remains weak.
“The pace of AI innovation is outstripping our ability to regulate it,” noted Dr. Alan Chen, a biosecurity expert at the Center for Global Health Security. “We need international agreements that treat AI-designed pathogens as seriously as nuclear weapons.”
A Double-Edged Sword
The same technology that can create therapeutic phages could, in theory, be used to engineer more virulent viruses. This dual-use dilemma is not new, but AI amplifies the risk because it lowers the barrier to entry—what once required a sophisticated lab can now be done with a laptop and an open-source model.
Researchers are calling for responsible innovation: embedding safety features into AI models, such as filters that block requests for known pathogens, and implementing mandatory screening of synthetic DNA orders. Companies like Profluent say they are committed to safe practices, but they also acknowledge that policy must evolve in tandem with technology.
Looking Ahead
As AI continues to advance, the line between beneficial and harmful applications becomes increasingly blurred. The creation of these 16 viruses is a testament to human ingenuity, but it also serves as a stark reminder that with great power comes great responsibility. In 2026, the scientific community stands at a crossroads: harness AI for good while safeguarding against its misuse, or risk unleashing consequences we cannot control.
For now, the researchers emphasize that their work is intended to spur dialogue and inspire cautious optimism. “We are opening a door to a new era of medicine,” said Dr. Carter, “but we must walk through it with our eyes wide open.”
This article is based on reporting from wired.com and includes context from the 2026 biosecurity landscape.
via Wired AI