Policy and governance limits of procurement as AI regulation

Policy and Governance Limits of Procurement in Military AI Regulation

The rapid development of artificial intelligence (AI) technologies has prompted significant discussions around effective governance mechanisms, especially in the context of military applications. Traditionally, procurement-based strategies—relying on contracts and purchasing agreements—have served as primary tools for regulating military AI. However, recent analyses highlight both their strengths and fundamental limitations, prompting calls for broader, more comprehensive policy frameworks.

Analyzing Procurement as a Governance Tool

Procurement, through contracts and purchasing regulations, offers a straightforward method for governments to influence military AI development. By setting requirements, standards, and oversight within procurement processes, authorities aim to ensure that AI systems meet safety, ethical, and strategic criteria. This approach allows for direct control over specific projects and facilitates accountability, making it an attractive governance mechanism in a complex technological landscape.

However, as Jessica Tilli's Lawfare critique emphasizes, reliance solely on contracts to regulate military AI is inherently limited. Contracts tend to focus on compliance with predefined standards and deliverables, but they often lack the flexibility to address rapidly evolving AI capabilities and emerging ethical concerns. Moreover, procurement contracts primarily regulate what is bought, rather than how AI is developed, deployed, or controlled once in operational use. This makes it challenging to enforce responsible AI practices across all facets of military development, especially when adversaries may operate outside formal procurement channels.

The Case for Broader Cooperative Policy Mechanisms

Given these limitations, experts advocate for broader strategic cooperation frameworks that extend beyond procurement. A recent RAND report on AI cooperation underscores the importance of core functions that facilitate responsible development and deployment of military AI. These core functions include:

• Information sharing on best practices and emerging threats
• Joint standards and norms for AI safety and ethics
• Collaborative research to develop shared technologies
• Mutual confidence-building measures to prevent escalation and misunderstandings

By fostering international and interagency cooperation, these mechanisms aim to create a more resilient and adaptive governance landscape. Such cooperation can help mitigate risks associated with AI proliferation and ensure that military AI systems align with shared strategic and ethical principles.

Balancing Strengths and Limitations

While procurement provides a valuable tool for immediate oversight, its limitations necessitate supplementing it with broader policy mechanisms. These include international treaties, multilateral agreements, and collaborative research initiatives that can adapt to the fast-paced evolution of AI technology. Implementing a layered governance approach enhances transparency, accountability, and ethical standards, reducing reliance on contracts alone.

Conclusion

The governance of military AI through procurement offers tangible benefits in terms of control and accountability but falls short in addressing the full spectrum of challenges posed by advanced AI systems. As Jessica Tilli’s critique and the RAND report highlight, effective regulation requires a combination of contractual oversight and broader cooperative policies. Building robust, flexible, and internationally coordinated frameworks is essential to ensure the responsible development and deployment of military AI, safeguarding strategic stability and ethical standards in this rapidly evolving domain.