Human-Centered Leadership is Key to Government’s AI Evolution

The future of artificial intelligence in government will depend less on the technology itself and more on how effectively people and institutions adopt and implement it, Navy CTO and Georgetown University professor Justin Fanelli said Monday.

“We are, from an AI perspective, still in very early days,” Fanelli said during a fireside chat with Hupside Managing Director Jonathan Aberman at Modev’s GovAI Summit in Arlington, Virginia. “Many government organizations aren’t necessarily getting the return on investment that they need to.”

Fanelli described the current stage of AI adoption in government as “Horizon Two,” referencing McKinsey’s framework for innovation maturity. While billions are being spent, he said, only a fraction of organizations report satisfaction with their outcomes.

“We are in a K-curve, frankly, for people and in a river that we will see not slow down,” said Fanelli. “Everyone spends the same on AI, but they’re not going to get the same results.”

Aberman noted that government leaders must better understand management principles to grasp AI’s broader effects — on the workforce, on agency operations and on how success is measured.

“We’re going to ultimately need to change the way that we do business, which means you’re going to have to convince non-believers — which means you need data stories,” Alberman said. “The people parts and that scorecard are the way that we get over the hump.”

Fanelli agreed that success in AI adoption hinges on leadership and incentive structures, not just new tools. Too few public sector leaders, he said, are thinking clearly about how to measure AI’s impact.

“If we can show for every billion dollars we spend on AI what we’re getting for it, I don’t think anything stops,” he said. “It’s a conversion problem … that is a people problem.”

He also challenged common narratives around AI’s disruption of the workforce, arguing that jobs will evolve rather than disappear. Agencies, he said, must focus on training and personalization to help employees adapt and innovate.

“There are people who will be disrupted, and there are people who will disrupt themselves,” he said. “Jobs will continue to be there, but the definition of jobs will change.”

Aberman compared AI’s rise to other transformative moments in human history, noting that societies that best leverage new tools tend to advance the fastest.

“AI is a tool. People make tools. If you look at the history of humanity since the development of agriculture, the groups of humans that supplanted or succeeded were the ones that used the current tools the best,” Alberman said. “It’s no different. The society that uses AI best as a society will dominate.”

Fanelli closed with a call for urgency and focus as government agencies navigate the challenges of AI transformation.

“You can sit at home and do doomsday planning,” Fanelli said. “Or you can get really good at important stuff.”