White House Curbs High-Risk Life Sciences Research, Citing AI-Driven Biosecurity Threats
Updated
Updated · GovInfoSecurity.com · Aug 14
White House Curbs High-Risk Life Sciences Research, Citing AI-Driven Biosecurity Threats
3 articles · Updated · GovInfoSecurity.com · Aug 14
Summary
A late-July White House policy bars federal support for dangerous gain-of-function research in the U.S. and abroad, adds independent review for certain high-risk life sciences work, and limits funding to institutions lacking adequate biosafety oversight.
The 12-page paper says little about AI directly but orders an interagency group to track AI-biology advances as experts warn frontier models and specialized biological tools could help design agents that evade vaccines, immunity, and gene-synthesis screening.
RAND researchers said AI is lowering the expertise barrier by turning parts of lab know-how into software-guided tasks, though experts stressed creating a novel pathogen still requires deep biological knowledge, testing, and specialized equipment.
Concern is shifting from chatbots to autonomous AI agents: RAND said agents can already navigate high-risk biology tools and code repositories, while open-source releases pose lasting risks because more than 60% of the highest-risk models it studied were fully open source.
Experts said effective safeguards must extend beyond AI developers to the full research stack—securing biological data and models, tightening access controls, mandating gene-synthesis screening, and building early-warning and response systems without choking beneficial medical research.