Without safeguards, AI-Biology integration risks accelerating future pandemics

TL;DR

AI tools for protein design could speed up biological research but also pose risks for creating dangerous viruses if not properly controlled.

Contribution

The paper introduces a framework for evaluating and safeguarding AI-biology systems to mitigate dual-use risks.

Findings

pLMs can optimize proteins and accelerate biological design, but also enable risky viral evolution experiments.

Current AI-biology systems lack safeguards for unintended or malicious use in laboratory settings.

A capability-oriented framework is proposed to guide safe development and evaluation of AI-driven biological tools.

Abstract

Artificial intelligence now shapes the design of biological matter. Protein language models (pLMs), trained on millions of natural sequences, can predict, generate, and optimize functional proteins with minimal human input. When embedded in experimental pipelines, these systems enable closed-loop biological design at unprecedented speed. The same convergence that accelerates vaccine and therapeutic discovery, however, also creates new dual-use risks. We first map recent progress in using pLMs for fitness optimization across proteins, then critically assess how these approaches have been applied to viral evolution and how they intersect with laboratory workflows, including active learning and automation. Building on this analysis, we outline a capability-oriented framework for integrated AI–biology systems, identify evaluation challenges specific to biological outputs, and propose research directions for training- and inference-time safeguards.

Figures