Loading page…
The company describes a physician-led treatment for an antibiotic-resistant infection. Its laboratory findings do not show whether the patient benefited.
Listen to this story
Radical Numerics says its Omnii-1-designed phage cocktail was administered to a patient with an antibiotic-resistant infection through compassionate use, extending AI-designed viruses from lab work into a human treatment setting. The company has not disclosed the patient’s clinical course, and says the cocktail killed the patient’s bacterial isolate only in pre-treatment lab tests; that does not establish benefit in the body. The account is self-published and not peer reviewed, so clinical efficacy remains unverified.
The physician-led compassionate-use treatment was not a clinical trial demonstrating that the approach works.
In a T7 experiment, 69.9% of designs with 11–15 amino acid changes produced viable phages, versus 4.8% of random controls.
Omnii-1 proposes edits to selected genome regions, while laboratory testing measures activity and guides subsequent design rounds.
AI-designed viruses have reached a human patient, according to Radical Numerics—but whether the treatment helped remains undisclosed. In an October 20 research account, the startup said viruses designed with its Omnii-1 model were administered against an antibiotic-resistant bacterial infection. The company describes a physician-led compassionate-use treatment, not a clinical trial demonstrating that the approach works.
Radical Numerics says the administration took place under an authorized compassionate-use pathway. That route allows patients with serious conditions to receive investigational therapies outside clinical trials when satisfactory alternatives are unavailable. It remains subject to oversight and safety requirements set by the relevant government authorities.
The treatment uses phages: viruses that infect bacteria. Phage therapy itself is not new, nor is the use of engineered phages in patients. The company's proposed advance is using AI to help design and select viruses suited to a particular infection, rather than relying only on a search through existing phage collections.
That matching problem begins with a sample of the bacteria causing the infection. Researchers screen phages against it, looking for candidates that can kill the bacteria. A phage's narrow targeting can be useful, but it also makes finding a suitable treatment difficult: a virus that infects one bacterial strain may fail against another.
Omnii-1 is a genome language model, trained on genetic sequences and further trained on phage genomes. Radical Numerics says it can generate replacement DNA for selected regions while accounting for the surrounding sequence. This does not require inventing an entire viral genome: the company describes changing regions that help determine which bacteria a phage can recognize.
The company also trains the model on measured interactions between phages and bacteria to predict compatibility. Its workflow then returns to the laboratory: researchers make the designed variants, measure their activity and use those results to inform further design rounds. The model guides which candidates to test; physical experiments supply the feedback.
In an illustrative experiment with T7, a phage that infects E. coli, Radical Numerics modified a protein involved in recognizing bacterial hosts. Among designs with 11–15 amino acid changes, it recovered 69.9% of AI-designed variants, compared with 4.8% of random controls. Recovery is a proxy for whether a design can produce viable phages—not a measure of treatment success.
For the patient-specific treatment, Radical Numerics says its phage cocktail eliminated the patient's bacterial isolate in laboratory testing before administration. That is a result against a sample outside the body, not evidence that the patient's infection cleared.
The company's account is self-published and has not been peer reviewed. Futurism reports that Radical Numerics declined to share clinical information about the patient or the doctors with the Times, citing the proprietary nature of a treatment it hopes to monetize.
Radical Numerics says it is withholding the patient's clinical course and outcomes to protect privacy. CEO Eric Nguyen told the Times that the company intends to publish its findings through peer-reviewed research, with future disclosure of clinical details subject to appropriate consent.
Story updates
AI-designed phages may offer another way to target antibiotic-resistant bacteria, but experts say a reported treatment is not proof of safety or effectiveness. The patient’s outcome has not been disclosed.
Getting AI-designed viruses into a patient does not establish that they make a safe treatment. In a November 13 article from Northeastern Global News, biotechnology experts warned about unreliable predictions and viral evolution after Radical Numerics said it had used AI-designed viruses against an antibiotic-resistant infection. The patient’s outcome remains undisclosed.
In its October 20 announcement, Radical Numerics said viruses designed using its Omnii-1 model were administered under an authorized, physician-led compassionate-use pathway. That allows patients with serious conditions to receive investigational treatments outside clinical trials when satisfactory alternatives are unavailable, subject to regulatory oversight and safety requirements.
The report has not been peer-reviewed. Radical Numerics cites patient privacy for withholding the clinical course and results; it also has not disclosed how the viruses were administered. Northeastern Global News says the treatment has not received FDA approval and still needs to go through clinical trials.
Northeastern biotechnology professor Sudhakaran Prabakaran described himself as cautiously optimistic, but warned that mutations as viruses reproduce could undermine patient safeguards. Alexander Sulakvelidze, president and CEO of phage developer Intralytix, flagged unreliable AI predictions, unintended effects and potential misuse to create harmful biological agents.
AI should complement, rather than replace, scientific expertise and laboratory testing.
Phages, short for bacteriophages, are viruses that infect bacteria. Engineered phages had already been used in patients. Radical Numerics’ proposed advance is AI-assisted design and selection, rather than relying only on searching existing collections for viruses that kill a patient’s particular bacteria.
A phage must recognize a compatible target on a bacterium’s surface. Researchers can alter the DNA encoding its binding proteins to change which bacteria it recognizes. That does not require inventing an entire viral genome: the company describes modifying selected regions while preserving the broader genome.
Radical Numerics says Omnii-1 predicts compatibility between phages and bacterial strains to prioritize experiments. Attachment is only the first hurdle: a phage must deliver its genetic material, overcome bacterial defenses and reproduce. The company says it makes and tests model-designed variants, using laboratory results to inform later designs.
Personalization also leaves a regulatory question unresolved. Prabakaran asked whether a different virus for each patient would require a fresh FDA sign-off each time. “All of these questions are open,” he said.
Loading discussion...
Join the conversation
Explain where you would draw the line between useful disclosure and premature promotion.
Be the first to share a perspective or an experience.
Reader comments
Newest comments first. Replies stay oldest first.