BioVeil MATRIX: Uncovering and categorizing vulnerabilities of agentic biological AI scientists

TL;DR

BioVeil MATRIX introduces a taxonomy to categorize biosecurity risks of agentic biological AI scientists, highlighting vulnerabilities and proposing safeguards to prevent dual-use applications in life sciences.

Contribution

The paper presents BioVeil MATRIX, a comprehensive taxonomy for biosecurity risks in AI scientists, and demonstrates increased capabilities and vulnerabilities in existing models like Biomni.

Findings

Agentic AI scientists can assist with dual-use biological tasks.

Current safeguards are insufficient against certain agentic behaviors.

BioVeil MATRIX provides a structured framework for risk assessment.

Abstract

Agentic AI scientists equipped with domain-specific tools are rapidly entering scientific workflows across disciplines, with especially strong uptake in the life sciences where they can be used for literature synthesis, sequence analysis, and experimental planning support. While these systems accelerate biological research, they also introduce risks for dual-use applications that are not captured by current model-centric safety evaluations. We present evidence that current agentic AI scientists, including Biomni and K-Dense, are willing to assist with dual-use tasks that are blocked by base model safeguards. We also found that in a paired evaluation framework for biology and chemistry prompts involving Weapons of Mass Destruction proxies (WMDP), agentic scaffolding of Biomni increased the benchmark performance relative to the underlying standalone model, producing measurable capability uplift. We believe it is necessary to include additional safeguards in existing models and build future tools from the ground up with agentic vulnerabilities in mind. To systematically categorize broader risks, we introduce BioVeil MATRIX, a defensive taxonomy that maps AI-enabled biosecurity risks using 10 tactical categories (TA01--TA10) and 22 different techniques. We propose to use this taxonomy as a baseline for future AI scientist development and generate specialized benchmarks and protocols for red-teaming these vulnerabilities before public deployment. BioVeil MATRIX can be found at: https://bioveilmatrix.com/