If there was a single unifying theme at the recent ARMA Public Sector InfoGov Summit 2026, it was this: Artificial intelligence will only be as effective as the information governance beneath it.
The event brought together leaders from the White House, the Department of the Interior, the National Archives, the VA and AMTRAK. What stood out was not enthusiasm for automation alone, it was a shared recognition that governance must mature alongside AI.
Philip Droege’s keynote on White House records management grounded the conversation in accountability. Managing presidential records across administrations requires precision, continuity and transparency. The lesson for the broader public sector is straightforward: governance frameworks are not optional infrastructure. They are the backbone of institutional trust.
As agencies adopt AI, that backbone becomes even more critical.
John Montel’s session on preparing the information environment for AI made this explicit. Before agencies introduce advanced analytics or generative tools, they must address ownership gaps, inconsistent retention practices and unreliable metadata. AI cannot compensate for disorganized information. In fact, it magnifies weaknesses.
The National Archives offered a forward-looking but disciplined example. Jill Reilly described how NARA is piloting AI-assisted description and semantic search to expand access to historical collections. However, the success of those pilots depends on solid metadata, trusted datasets and transparent implementation. AI needs to introduced incrementally and evaluated rigorously.
This approach reflects a broader shift in mindset.
Government is moving from digitization as the goal to intelligent activation as the next step. Digitizing records was the first phase of modernization. The next phase uses AI to extract structure, enhance discoverability and surface insight but always within auditable and defensible frameworks.
FOIA provided another lens on this evolution. Dr. Moya Hill highlighted how transparency mandates intersect with privacy and records management. AI can help agencies respond to requests more efficiently by identifying related documents and standardizing redactions. Yet transparency requires explainability. Agencies must be able to demonstrate how decisions were made and ensure human review remains integral.
Perhaps the most forward-looking discussion centered on the concept of a “governance fabric,” introduced in the closing keynote. Information governance can no longer reside in isolated departments. It must connect IT, records, data and business units. As AI systems become more capable potentially coordinating workflows or flagging compliance risks autonomously governance rules must travel with the information itself.
This Is the Bridge to Agentic AI
In practical terms, agencies are progressing through three stages.
First, intelligent extraction, using AI to classify and structure incoming content.
Second, human-plus-AI collaboration, accelerating review and surfacing insights while preserving human judgment.
Third, intelligent orchestration, where AI can monitor thresholds, recommend next steps and support predictive insights within clearly defined guardrails.
At each stage, the same principle applies: AI augments people. It does not replace accountability.
The agencies represented at the Summit are not pursuing AI as a standalone initiative. They are aligning it with records policy, cybersecurity priorities and digital modernization strategies. They are building cross-functional partnerships between IT and records leaders. They are investing in metadata quality before expanding automation.
The message for government leaders is clear.
Modernization is no longer about simply migrating systems or scanning documents. It is about creating an information environment that is structured, trusted and governed so AI can operate responsibly within it.
Agencies that approach AI deliberately with governance as the starting point rather than an afterthought will reduce risk, increase transparency and improve service delivery simultaneously.
The future of AI in government will not be defined by the sophistication of the algorithms alone.
It will be defined by the strength of the governance fabric that surrounds them.
Andy MacIsaac is a senior marketing leader at Laserfiche, where he drives go-to-market strategy and thought leadership for AI-powered content management, process automation, and data governance in the public sector. With more than two decades of experience partnering with government agencies and education institutions, he helps organizations modernize operations while maintaining security, compliance, and trust. Andy has led industry marketing, demand generation, and sales enablement initiatives across leading software and consulting organizations, translating complex technologies into practical outcomes. As a trusted advisor to CIOs and agency leaders, he is passionate about responsible innovation that improves efficiency, transparency, and service delivery.
Leave a Reply
You must be logged in to post a comment.