---
title: Biosafety
type: vocabulary
url: "https://www.envisioning.com/vocab/biosafety"
summary: Containment principles, practices, and protocols to prevent unintended harm from biological agents.
year: 1970
generality: 0.55
---

# Biosafety

Containment principles, practices, and protocols to prevent unintended harm from biological agents.
Biosafety is the discipline of containment principles, laboratory practices, equipment design, and facility architecture intended to prevent unintentional harm from biological agents to laboratory workers, the surrounding community, and the broader environment. The term and the practice have a long history stretching back to the recognition that pathogens could be handled safely in research settings, formalized in the 1970s and 1980s through the Biological Safety Level (BSL-1 through BSL-4) classification system used internationally and through national and international guidelines for pathogen work. Biosafety sits alongside biosecurity, which addresses intentional misuse of biological agents, and the two are increasingly discussed together as dual-use concerns in the AI × bio era.

The mechanism of biosafety is layered: engineering controls (biological safety cabinets, sealed facilities, dedicated ventilation systems, autoclaves) form the first and most reliable layer of protection; administrative controls (training, standard operating procedures, medical surveillance) form the second; and personal protective equipment forms the third. Each layer addresses what would happen if the layer above it failed. The BSL classification assigns a level of containment to a pathogen based on its hazard, ranging from BSL-1 (well-characterized agents not known to cause disease in healthy adults) through BSL-4 (dangerous and exotic agents with no available treatment, such as Ebola and Marburg viruses). Work at higher containment levels requires facilities, equipment, and training that are correspondingly more stringent.

The tradeoffs in biosafety are predominantly between safety and pace of research. The most stringent containment reduces risk at the cost of slowing experimental work, increasing the expense of any given experiment, and limiting the breadth of questions that can be asked. Since 2010, biosafety has also intersected with biosecurity, with the rise of gain-of-function research — experiments that increase the transmissibility or virulence of pathogens — and with the emergence of synthetic biology and now AI-designed biology, which lowers the technical barrier to producing novel constructs. The combination has produced national-level biosafety policy frameworks in the United States (the dual-use research review framework, the NIH P3CO framework), in Europe (comparable pathogen-by-pathogen reviews), and at the WHO. Public and political trust in these frameworks has fluctuated as high-profile controversies around specific research programs have emerged.

Open questions include whether current biosafety frameworks are appropriate for AI-designed biology, where the identity of the engineered construct may not be straightforwardly mapped to existing risk categories, and how to govern research distributed across many countries with varying legal regimes. The connection to AI is direct and growing: AI tools for biological design expand the population of researchers who can produce novel biological constructs, and raise the question of whether existing biosafety frameworks — written for human researchers — adequately cover AI-driven design pipelines whose outputs may include engineered organisms with properties outside the current risk classification. As generative biology, autonomous wet labs, and AI-designed biology scale up, biosafety frameworks will need to keep pace.

---
Source: Envisioning — Technology Research Institute (https://www.envisioning.com/vocab/biosafety)