Strengthening global biosecurity for synthetic nucleic acid technology: from sequence screening to risk-based governance in the AI era

Abstract

Synthetic nucleic acid technology has expanded rapidly, reducing the cost, time, and technical barriers associated with DNA and RNA synthesis. Existing governance frameworks have focused mainly on sequence-based screening of synthesis orders and customer verification. Although these measures remain essential, they are no longer sufficient as standalone safeguards in a landscape shaped by AI-assisted biological design, globally distributed synthesis capacity, uneven regulatory implementation, and unclear liability across the design-synthesis-use pathway. Protein language models, generative design tools, and other AI-assisted approaches may expand the sequence design space in ways that are harder to assess through homology-based screening alone. This review examines the current biosecurity architecture for synthetic nucleic acid technology and identifies structural limitations in sequence-centered governance. We propose a systems-based, risk-informed framework organized around four interacting dimensions: design capability, access, governance strength, and liability and attribution clarity. The framework is diagnostic and semi-structured rather than quantitative, and is intended to complement existing screening standards and national regulatory approaches. Building on it, we present a four-tier governance escalation model that links dimension profiles to proportionate policy responses and assigns responsibilities to synthesis providers, research institutions, governments, funders, and international bodies. A worked case study illustrates how the framework can be applied to a plausible governance scenario with growing synthetic biology capacity and uneven oversight. The approach integrates equity and proportionality as governance constraints, supporting biosecurity strengthening without undermining responsible innovation.