GC REAIM Expert Policy Note Series

Over the past two weeks, we’ve launched the GC REAIM Expert Policy Note Series, sharing six of the 19 thought-provoking policy notes written by members of the GC REAIM Expert Advisory Group. These notes explore the variety of issues that emerge at the intersection of AI, ethics, law, and military strategy, offering timely, practical recommendations for responsible AI governance.

Here’s what we’ve covered so far:

Policy Note #1: Effective Governance Through Precise Common Understanding
Authored by Edson Prestes, Nayat Sanchez-Pi, and Maria Vanina Martinez, this policy note emphasizes the importance of clear definitions and structured taxonomies in understanding AI’s capabilities and limitations. In the military context, avoiding ambiguity is crucial for effective communication and interoperability across various stakeholders.
Read the full note here

Policy Note #2: AI and Nuclear Stability
In this policy note, James Johnson delves into the risks of integrating AI into military and, specifically, nuclear systems. He highlights how AI could increase escalation risks in the context of nuclear deterrence and offers a framework for mitigating these risks, including global safety standards and enhanced human oversight.
Read the full note here

Policy Note #3: The Risks of Integrating Generative AI into Weapon Systems
Vincent Boulanin explores the potential risks of using generative AI in weapon systems, including accidental harm, misuse, and the erosion of human control. The policy note calls for cautious adoption and outlines risk mitigation strategies, including technical, organizational, and policy measures.
Read the full note here

Policy Note #4: Applying International Law to AI in the Military Domain

Written by Fan Yang, this note proposes an integrated legal approach to addressing AI-specific risks in military contexts. It evaluates how existing international laws can be applied to AI-driven lethal autonomous weapons and cyber operations and calls for nuanced, context-specific legal analysis.

Read the full note here

Policy Note #5: Collective Moral Responsibility and LAWS
In this paper, Seumas Miller examines moral and institutional responsibility surrounding the use of lethal autonomous weapon systems (LAWS). He stresses the necessity of meaningful human control and argues that accountability must remain with human actors, not AI systems.
Read the full note here

Policy Note #6: Restrictions on AI Weapons in Specific Situations
Mun-eon Park explores when and how the use of autonomous weapons could be restricted under international humanitarian law. The note addresses legal and ethical concerns around human dignity and emphasizes the importance of human control before, during, and after attacks.
Read the full note here

Stay tuned as we continue publishing the rest of the policy notes throughout April and May, each bringing critical insight into how we can responsibly shape the future of AI in the military domain.

All policy notes will be available on the GC REAIM website.