GovLoop

How Data Management Drives AI

Technology helps agencies work more efficiently and effectively, and AI tools, in particular, are uniquely powerful. Whether generative AI (GenAI), natural language processing, machine learning (ML) or another AI option, these tools can transform how organizations engage with the public.

“The integration of AI within government operations will redefine our interaction between citizens and government,” said Chris Steel, AI Practice Lead with AlphaSix, which provides data management platforms and data analysis tools. “It will make it a lot more personalized, a lot more efficient overall.” AI-driven automation and analytics can both streamline processes and make government services more accessible and responsive, he said.

But the foundation of AI is data — high-quality, accessible and secure. Think of a pyramid, Steel said. The top 10% or 20% is AI. Everything below is data and data management.

Fabric, Not Stovepipes

It’s tough to manage data across large organizations. “If your data is spread all over the place in a bunch of stovepiped systems, it’s going to be hard to really unlock [its] true value … and then [be] able to deal with the volumes of [it],” said Stephen Moore, Chief Technology Officer, also with AlphaSix. People often summarize data or collect limited datasets because of staff constraints, he said, but then “they’re missing the input that you would need to really get the output … you’re looking for.”

The solution, Moore offered, is to create a data fabric — a standardized, secure way to integrate various data pipelines so the data can be accessed and analyzed easily. For instance, AlphaSix helped a large federal agency build a centralized, protected system that each day brings in billions of records from security, fraud and other applications. Analysts can access, query and evaluate data they otherwise couldn’t, said Moore.

Agencies also can incorporate AI and ML models to identify precursors of negative events based on how prior bad events played out.

Churning Through Data

Analysts say they spend 80% of their time just getting access to the data, preparing it for analysis and configuring the tools, and they spend only 20% of their time actually doing whatever they’re supposed to do, Moore noted.

The amount of data they must sort through is probably 1,000 times what a person can handle, he said. But AI not only can process far more data, and faster, it can anticipate issues humans cannot. AI tools “don’t need to sleep [and] they don’t need to eat,” Moore said. “They can just sit there 24 hours a day churning through the data.”

Skilled AI professionals are scarce, however, and that makes it more difficult to integrate AI into legacy technology and weed out data biases. That’s when a partner such as AlphaSix comes in — as the AI expert when agencies cannot hire their own, said Steel.

The technology can transform government, he explained, “but data management is the biggest part of the AI problem.”

This article appears in our guide, “Leveling Up With Data.” To read more about how agencies are upping their data game, download it here:

Photo by Christina Morillo at pexels.com
Exit mobile version