Biosecurity Threats
The potential misuse of general purpose AI models also extends to biosecurity threats. Biological weapons are generally understood as biological toxins or infectious agents such as viruses that are intentionally released to cause disease and death.157 General purpose AI models could facilitate the production of biological weapons, by reducing barriers through access to critical knowledge or increasingly automated assistance and thus enable more malicious actors.
ENTITY
1 - Human
INTENT
1 - Intentional
TIMING
2 - Post-deployment
Risk ID
mit842
Domain lineage
4. Malicious Actors & Misuse
4.2 > Cyberattacks, weapon development or use, and mass harm
Mitigation strategy
1. Mandating rigorous, independent pre-deployment model audits and comprehensive biosecurity risk assessments to evaluate the AI model's capacity to lower the technical barriers for the design, production, or release of dangerous biological agents. 2. Implementing and continuously refining technical guardrails, such as robust input and output filtering and runtime monitoring, on general-purpose AI models to block queries and generation of step-by-step guidance or critical knowledge for the synthesis and deployment of high-consequence biological threats. 3. Establishing and enforcing international governance frameworks and domestic legislation that require enhanced supply-chain regulation for biotechnology materials, equipment, and sensitive datasets, specifically addressing the dual-use nature amplified by AI capabilities to restrict malicious access to physical resources.