What's Happening?
A February 2026 study by the Stockholm International Peace Research Institute (SIPRI), supported by a Department of Defence grant, emphasizes that procurement processes can integrate responsible-AI commitments into practice for military AI systems. This
is contingent on states maintaining an independent capacity to assess supplier claims. The study also notes that for systems receiving continuous updates, evaluation must extend beyond the initial purchase. Canada is rapidly acquiring military robotics and AI-enabled systems, with Lieutenant-General Darcy Molstad, commander of Canadian Joint Forces Command, indicating a steady stream of requests for proposals over the next five years. While Molstad stressed responsible acquisition with appropriate policies and checks, the SIPRI study underscores the importance of contractual safeguards. These safeguards should grant officials access to test results and incident reports, mandate suppliers to disclose significant changes, and specify when retesting is required. Additionally, contracts should ensure continuity of support, enabling Canada to investigate failures and determine the suitability of updated systems.
Why It's Important?
The SIPRI study's findings are crucial for ensuring the responsible and effective integration of artificial intelligence into military operations. Without robust procurement processes that include independent assessment and continuous evaluation, there is a risk of acquiring systems that may not align with ethical guidelines or operational requirements. The reliance on private companies for the underlying infrastructure of military AI also raises concerns about control and potential vulnerabilities. If Canada and its allies do not establish common standards for evidence and testing, warnings about malfunctions could be overlooked, potentially compromising military effectiveness and safety. This issue is particularly pertinent as military AI development outpaces diplomatic efforts to establish international norms and safeguards. The study highlights the need for Canada to define its 'red lines' regarding the use of lethal autonomous weapons and champion civilian protections in international discussions.
What's Next?
Canada's military effectiveness will increasingly depend on the choices it makes regarding the procurement and use of AI systems. The SIPRI study suggests that more testing and research are needed to provide Canadian diplomats with evidence for international negotiations. Canada could collaborate with partners to establish common requirements for suppliers and test whether military personnel can effectively challenge AI recommendations under pressure. This could lead to common purchasing requirements and international safeguards. Domestically, Canadian military exercises could be designed to introduce flawed AI recommendations to examine how personnel respond and whether interventions change decisions. Through initiatives like the Responsible AI in the Military Domain (REAIM) process, Canada could work with international partners to develop common tests for military decision-support systems, using unclassified scenarios to compare responses to conflicting information or declining system reliability. This research could inform Canada's contribution to international efforts to understand human judgment and control over military AI.
Beyond the Headlines
The implications of the SIPRI study extend beyond immediate procurement practices, touching upon deeper ethical, legal, and cultural dimensions of military AI. The question of control, particularly when private companies own the infrastructure and allies may apply different standards, presents a complex challenge to national sovereignty and operational autonomy. The study also implicitly raises concerns about the 'human in the loop' and the potential for human judgment to be compromised under pressure or by flawed algorithmic advice. The need for technical expertise within the Canadian military to investigate failures and assess updated systems underscores a broader requirement for workforce development and training in AI. Furthermore, the call for regulatory frameworks and channels for information exchange, confidence-building, and crisis clarification highlights the critical role of international diplomacy in mitigating the risks associated with advanced military technologies. The development of common international standards for military AI could set a precedent for global governance in other emerging technological domains.













