DigestAI news desk

Cut through the AI noise.

Policy & Regulation3 min read

Opinion: Radical Numerics warns biosecurity AI arms race may outpace defense

In an interview with Latent Space, Eric Nguyen, co‑founder and CEO of Radical Numerics, argues that advances in genomic language models (GLMs) are turning bio‑security into an AI arms race. He notes that early models like Evo and Evo‑2, developed at the Arc Institute and Stanford, were already capable of generating entire bacteriophage genomes that could be synthesized into functional viruses.…

1 source

Key points

  • Radical Numerics co‑founder Eric Nguyen says GLMs can generate full virus genomes, raising bio‑security concerns.
  • Long‑context models, first appearing three years ago, now handle DNA sequences up to 3 billion bases.
  • Nguyen’s experiment showed a GLM could extrapolate higher‑scoring RNA aptamers from lower‑scoring examples.

Nguyen says the same technologies that boost biological research also give malicious actors new tools, and he believes current defense efforts are falling behind. He describes experiments where a GLM was trained on low‑scoring RNA aptamers and then asked to extrapolate higher‑scoring sequences, successfully reproducing results it had not seen. The interview frames bio‑security as a race where defensive AI must accelerate to keep up with offensive capabilities.

The discussion highlights the need for open, defensive AI research to match frontier model capabilities, echoing calls from Anthropic and others to treat bio‑security with the same urgency as cyber‑security.

Full story from Latent Space · by RJ HonickyOpen source ↗

🔬Bio-security is an AI Arms Race - Eric Nguyen (CEO, Radical Numerics)

Latent Space · 23 September 2026

Loading the full article…

This text was published by Latent Space and written by RJ Honicky. It is reproduced here with attribution so you can read it in full; the rights remain with the publisher. Read it at the source ↗

Topics · follow one to build your own front page
Radical NumericsOpenAIHugging FaceAnthropicArc InstituteStanfordEvoEvo-2Eric NguyenClem DelangueMichael PoliStefano MassaroliArmin W. Thomas

The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us. Published by Martin K., who runs Digest AI and handles corrections.

Comments

via GitHub Discussions

More in Policy & Regulation

All →

Related stories