A biotechnology company says it has used artificial intelligence to design bacteriophages--viruses that target bacteria--and that the experimental treatment has been administered to a patient with a drug-resistant infection. Whether it helped the patient has not been disclosed.
AI-assisted targeting of bacteria
San Francisco-based Radical Numerics developed Omnii-1, an AI system that analyzes biological sequences to help design bacteriophages and predict which bacteria they may infect. Unlike viruses that target human cells, bacteriophages attack bacteria, making them a potential tool against infections that no longer respond to some antibiotics.
The company says its experimental phages eliminated bacteria isolated from the patient's infection in laboratory tests. The treatment was given through a physician-led compassionate-use pathway. However, the company has not shared the patient's clinical outcome, and its reported findings have not undergone peer review. Lab results alone cannot establish whether a treatment is safe or effective in a person.
Promise requires clinical evidence
Phages often target specific bacterial strains, which can make finding or designing a suitable match challenging. AI may help researchers explore possible designs more quickly, but each candidate still needs careful testing. Potential concerns include inflammatory reactions, effects on other bacteria, and unwanted genetic changes in bacteria. A 2026 consensus guideline in Nature Medicine highlighted the importance of product quality, patient monitoring and stronger clinical studies for phage therapies.
AI-designed bacteriophages could eventually add a precision tool to the response to antibiotic resistance, if rigorous clinical research confirms that they improve patient outcomes.
