Radical Numerics says its offline model identifies AI-engineered pathogens

single source· 1 articles · confidence: low · first seen 2026-11-11 17:00 UTC

What this means for you

Nothing to act on. There is no model to test, no evaluation with a date and harness, and no stated availability. A tool like this is judged on its error rates against real samples, and none have been published.

Radical Numerics, a startup, says it has built a model that can identify pathogens — natural ones and ones engineered with AI — without an internet connection. That is the whole of what Wired's report gives: no accuracy figures, no evaluation date, no statement of what the model was trained on, no release date and no price. That the model can do this is the company's claim; no published test is cited in support. Detection tools of this sort stand or fall on false-positive and false-negative rates against real samples.

Key facts

  • ·Radical Numerics is the startup named as developing the model. source
  • ·The company says the model can identify both natural and AI-engineered pathogens. source
  • ·The model is described as working without an internet connection. source
  • ·The report gives no accuracy figures, evaluation date, release date or price for the model. source

What the sources say

  • Wired AI — Brief report on an offline pathogen-detection model, with no accuracy data, release date or evaluation given.

Sources

The original reporting. Follow these — they did the work.

← the wire