How Federal Agencies Can Achieve AI Readiness Through Open Data and Strategic Data Management
AI Readiness in Government: Why Open Data Is the Key to Successful AI Adoption
In March 2024, the Office of Management and Budget dropped what’s widely being called the most significant AI policy guidance the federal government has ever issued: OMB Memo M-24-10. It lays out the rules of the road for how agencies should use artificial intelligence—how to manage risk, ensure transparency, and track usage. It’s a big deal.
But here’s the twist: the memo that will matter most in determining how successful agencies are at using AI wasn’t issued this year. It came out over a decade ago.
That’s right. I’m talking about OMB Memo M-13-13, cue the Jamaican air horn, the one that kicked off the federal open data policy and required agencies to start managing their data as a strategic asset. Back then, it was about transparency, innovation, and public access. Today, it turns out, it’s the groundwork for effective AI adoption.
Building AI Readiness Long Before We Called It That
Back in the early 2010s, governments across the world were embracing open data. Datasets started popping up online. APIs were built. Agencies were told: Don’t just keep your data in a vault—share it. The Obama administration issued an Executive Order, and M-13-13 followed soon after to set the standards.
The thing is, you can’t publish open data without doing some serious internal housecleaning first. You need to know where your data is. You need to ensure it’s accurate, complete, up-to-date, and secure. You need to document it well enough that people who didn’t write it can still understand and use it. Sound familiar?
These are the exact same qualities that make data usable for AI.
So whether we realized it or not, the agencies that took open data seriously over the past decade have been doing altitude training. They’ve built the organizational muscles, habits, and systems needed to handle data with care and clarity. And now, with AI suddenly demanding high-quality, well-structured data? These agencies are ready to sprint.
Why High-Quality Data Is the New AI Currency
Here’s the hard truth about AI tools—especially generative AI tools that everyone’s excited about right now: they’re only as good as the data they’re trained on.
If your data is messy, fragmented, outdated, or undocumented, it doesn’t matter how fancy your AI tool is. You’re going to get low-quality outputs. Worse, you’ll waste time and money trying to make sense of results that don’t actually serve your users.
The best results come from training these models on domain-specific data that reflects the work your agency actually does. And if you want to enhance that with techniques like Retrieval-Augmented Generation (RAG), you need seamless access to high-quality, structured information. Without that, the AI just spins its wheels.
Right now, the Department of Commerce is leading an effort to make open data more AI-ready—and that work is drawing a straight line from open data maturity to AI success.
What You Can Do Now
If you want to make the most of generative AI tools in government, look backward before you look forward. Ask yourself: How mature is our open data program? Have we built the processes, systems, and culture to support clean, accessible, useful data?
The agencies that have done this work—painstakingly, iteratively, often without the buzzwords—are already ahead. They’re not just ready for AI. They’ve been training for it.
My Role in This
Over the years, I’ve worked with public agencies to improve the way they manage and share data—not just to comply with policy, but to better serve people. Now, as AI becomes the next frontier in government tech, I’m helping those same agencies rethink their data through a new lens: How can we make this usable for the tools of tomorrow?
If you’re navigating the AI landscape and want a partner who can help you turn your data into a strategic asset—something not just compliant but powerful—I’d love to talk.
Let’s build systems that are as smart as the people they serve.
