Radical Numerics launches offline AI model to detect engineered pathogens
Radical Numerics unveiled Omnii Defense, an edge AI model meant to detect natural and AI-engineered pathogens using a laptop paired with a USB gene sequencer, and able to operate without internet access. The San Francisco startup, founded in 2025 by former Stanford researchers behind the Evo DNA models, previously used a related model called Omnii to design bacteriophages that treated a drug-resistant E. coli infection under FDA permission.
GoKawiil's interpretation of the reporting above, not reported fact.
CEO Eric Nguyen frames the move as an early step in what he sees as a looming arms race between AI-enabled bioweapon threats and AI-based defenses, drawing a parallel to cybersecurity's offense-defense dynamic. The announcement follows scrutiny of Radical after it did not disclose a partner, Tabula Bioscience, in its earlier phage research, a lapse that reportedly prompted adviser George Church to resign, which could raise questions about transparency as biodefense AI claims gain attention.
- Omnii Defense is designed to run offline on a laptop with a USB gene sequencer to flag dangerous pathogens.
- Radical Numerics' earlier model, Omnii, was used to create bacteriophages that treated a drug-resistant E. coli infection with FDA permission.
- An undisclosed partnership with Tabula Bioscience in that earlier work led adviser George Church to resign from Radical.
Source: wired.com — Emily Mullin, 2026-11-11
Published there as: “A Startup Wants to Fight AI Bioweapons With AI”
Read the original report → The summary and analysis above are GoKawiil's own, written from reporting by the source above. Facts and quotes belong to the original publisher.