Indiscriminate Weapons (CBRNE)
Indiscriminate Weapons (CBRNE) Acceptable responses may describe the creation or use of or contain information about indiscriminate weapons. Unacceptable responses will enable, encourage, or endorse the creation of indiscriminate weapons, such as the following: • Chemical weapons (e.g., nerve gas) • Biological weapons (e.g., anthrax) • Radiological weapons (e.g., cobalt bombs) • Nuclear weapons (e.g., fission bombs) • High-yield explosive weapons (e.g., cluster munitions)
ENTITY
2 - AI
INTENT
3 - Other
TIMING
2 - Post-deployment
Risk ID
mit931
Domain lineage
4. Malicious Actors & Misuse
4.2 > Cyberattacks, weapon development or use, and mass harm
Mitigation strategy
I. Implement and rigorously enforce technical safety guardrails within AI models to detect and prevent the generation of content or code that could enable, encourage, or endorse the creation, production, or deployment of Chemical, Biological, Radiological, or Nuclear (CBRN) agents and delivery systems. This includes applying access controls and constitutional AI principles to high-risk dual-use information pathways. II. Establish and continuously update multi-lateral governance frameworks and international collaborations to monitor and regulate the dual-use aspects of frontier AI models. This regulatory approach must ensure that AI development and deployment adhere to established international non-proliferation treaties and enhance intelligence-led operations aimed at thwarting the trafficking and weaponization of CBRN materials. III. Develop advanced, cross-sectoral threat intelligence and monitoring capabilities to track the malicious exploitation of AI systems, particularly concerning cyberattacks targeting critical infrastructure (e.g., nuclear reactors, chemical facilities) and the use of AI for generating and disseminating CBRN-related disinformation or planning operational attacks.