The Biosecurity Handbook
Biological Security in the AI Era
Welcome to The Biosecurity Handbook
The same AI that accelerates drug discovery can accelerate the design of dangerous pathogens. Biology is getting easier to engineer, and governance has not caught up. Written for biosecurity professionals who need AI context and for AI safety researchers who need biological grounding.
Three questions drive this handbook: How do emerging technologies change biological risk? What governance frameworks exist and where do they fail? What does credible preparedness and response require?
This resource is continuously updated as new research emerges.
A Note on Dual-Use Content
This handbook addresses biological security risks. Technical details that could enable misuse are cited from peer-reviewed literature but not expanded upon. The focus is frameworks, governance, and risk assessment, not operational protocols.
Recent highlights
- 27 Oct California AB 1864 (2026) synthesis-screening mandate and scope limits (opens the source in a new tab) The statute attaches screening duties to California sales rather than federal funding, while freezing its mandate to the September 2024 framework text. In DNA Synthesis Screening
- 27 Oct Maiwald et al. preprint on pretraining data filtering in protein models (opens the source in a new tab) For open-weight biological models, pretraining data filtering raises the cost of recovering a capability rather than permanently removing it. In How AI Amplifies Biosecurity Risk
- 27 Oct Feldman et al. Science letter on agricultural pathogens in AI biosecurity (opens the source in a new tab) Crop and livestock pathogens carry high food-security and economic stakes that human-centered screening lists and model evaluations can miss. In Policy Frameworks for AI-Bio Convergence
- 23 Oct Anthropic describes what its independent evaluators can see and publish (opens the source in a new tab) Outside review is only as strong as its access and publication rights, so readers can now judge what this oversight can and cannot catch. In Red-Teaming AI Systems for Biosecurity Risks
- 22 Oct WHO guidance on consent, reuse, and benefit sharing for human genome data (opens the source in a new tab) Human-genome reuse needs clear authorization and benefit-sharing terms for affected people and communities. In Laboratory Biosafety and Biosecurity
Start Here
Read the risk frame first, then how AI changes biological risk.
- Executive Summary: Key findings and recommendations
- What Is Biosecurity?: Core concepts and scope
- AI as a Biosecurity Risk Amplifier: How AI changes biological risk
Then continue to Policy Frameworks for AI-Bio Convergence for governance context.
For role-specific reading paths, see the Preface; for teaching, see the Course Syllabus Template and Case Studies in Biosecurity Failures and Successes.
An AI assistant helping you should follow the Guidance for AI Agents.
Companion Handbooks
The Physician AI Handbook
Peer-Reviewed Evidence Across Every Medical Specialty
What works. What doesn’t. What happens when AI is wrong.
The Public Health AI Handbook
Evaluating AI Tools for Public Health Practice
Some AI tools work in the field. Others only work on paper.
The Life Sciences AI Handbook
Evidence for AI in Biology and Drug Discovery
What works, what remains unproven, and what stands in the way.
Free to read and reuse under CC BY 4.0 · DOI 10.5281/zenodo.18252920 · How to cite
This handbook is open and continuously maintained. Support the Biosecurity Handbook →