Key Takeaways

• Federal AI policy is raising expectations for governance, security, and accountability. Agencies must demonstrate how AI systems are accessed, used, and controlled. 
• Identity enables AI governance in practice. Consistent identity controls support provenance, authorization, and accountability across systems. 
• Fragmented identity limits policy effectiveness. Gaps across systems reduce visibility, weaken enforcement, and introduce risk. 
• Proven ICAM frameworks like SADI provide a scalable path forward. They enable federation, real-time visibility, and consistent policy enforcement. 
• A unified identity foundation allows agencies to adopt AI securely and at scale. 

Federal agencies are moving quickly to adopt artificial intelligence to improve service delivery, automate workflows, and enhance decision-making. The Trump Administration’s National Policy Framework for Artificial Intelligence reinforces this momentum, positioning AI as a critical driver of national security, economic growth, and public trust. 

As agencies align with this policy direction, the focus is naturally shifting toward how to operationalize AI in a way that is secure, accountable, and scalable. One element sits at the center of that effort, even if it is not always called out directly, which is identity.

How does identity support AI governance?

Federal AI policy places a strong emphasis on accountability, transparency, and responsible use. In practice, that means agencies need to understand how AI systems are being accessed, what actions they are taking, and under what authority those actions occur. Answering those questions depends on identity.

When identity is consistently managed, agencies can trace access to data, validate system interactions, and enforce policy across AI-enabled environments. Identity becomes the mechanism that connects users, systems, and decisions in a way that is both verifiable and auditable.

Without that foundation, governance becomes more difficult to enforce, particularly as AI systems scale across agencies and mission areas.

What challenges do agencies face today?

Many agencies have made significant progress transforming systems, but identity often remains distributed across bureaus, programs, and applications. As AI expands and begins to connect data and workflows across those environments, gaps in identity become more visible.

Inconsistent identity controls can limit visibility into user activity, complicate policy enforcement, and introduce risk at the points where systems intersect. 

What may have once been a localized issue now carries broader implications as agencies move toward more integrated, AI-enabled operations.

Why does AI change the security equation?

AI introduces new layers of complexity that extend beyond traditional cybersecurity approaches. Systems are increasingly interconnected, data flows are more dynamic, and decisions may be automated at scale.

The federal AI framework highlights the need to address evolving risks, including misuse, fraud, and emerging threat vectors. 

In this environment, securing systems alone is not enough. Agencies also need to ensure that access is continuously validated, activity is visible, and policies are consistently enforced. Identity provides that continuity across systems, data, and AI models.

What does a scalable approach look like?

Rather than starting from scratch, agencies can build on proven identity frameworks that already operate at scale.

At the IRS, the Secure Access Digital Identity (SADI) framework, governed by the eA3 architecture, provides a model for how identity can function as an enterprise control layer. This approach supports centralized governance while allowing systems and organizations to maintain flexibility in how they operate.

By enabling secure federation, real-time visibility, and consistent policy enforcement, identity becomes an enabler of both security and innovation. It also extends naturally to non-human identities, which are increasingly important as AI systems and automated processes expand.

Why does this matter now?

As federal AI policy continues to evolve, expectations around governance, security, and transparency will only increase. Agencies are being asked to scale AI capabilities while maintaining trust, protecting data, and ensuring compliance.

A unified identity foundation makes that possible. It allows agencies to apply policy consistently, monitor activity across systems, and support secure, scalable AI adoption without slowing innovation.

AI is reshaping how federal agencies operate, but its success depends on the systems that support it.

Identity plays a central role in that foundation. When managed consistently, it strengthens governance, reduces risk, and enables agencies to move forward with confidence as they adopt AI at scale.If your agency is exploring how to align AI initiatives with federal policy or strengthen identity as a foundation for secure adoption, contact Makpar to learn how we help build resilient, mission-rea