AI is showing up in more workplaces every day, but for many government employees, it still raises a lot of questions. Is it safe to use? Will it change my job? How do I know when to trust the results?
Those are exactly the kinds of concerns supervisors should be talking about.
During our most recent Supervisors in Government Community of Practice session, we chatted with Dylan Hayden, Associate Professor of Practice, Fels Institute of Government, University of Pennsylvania, about how supervisors can lead productive conversations about using AI at work. Read on for a few highlights from that touched on setting AI expectations, encouraging responsible experimentation, addressing employee concerns, and identifying practical workplace use cases.
Start With the Work, Not the Technology
When teams are asked to implement AI, it’s easy to focus on the tool itself. But supervisors can have more productive conversations by focusing on day-to-day challenges instead.
Ask your team:
- What tasks take the most time?
- What’s causing frustration?
- Where are the biggest bottlenecks?
Starting with real workplace problems helps employees see AI as a potential helper, not just another technology initiative. It also shifts the conversation from “How do we use AI?” to “Can AI help us solve this problem?”
Acknowledge Concerns and Learn Together
Not everyone’s hesitation about AI comes from the same place. Some employees worry about job security. Others are concerned about privacy, security, or making a mistake that could have real consequences. Those concerns are understandable, especially in government environments where accuracy and accountability matter.
Rather than trying to have all the answers, supervisors can set a better example by being open to learning alongside their teams. Creating space for questions, sharing lessons learned, and encouraging responsible experimentation goes a long way toward building confidence.
Hayden commented that the goal isn’t to get people to blindly trust AI. The goal is to help them understand when to trust it, when to verify the results, and when human judgment needs to take the lead.
Make AI Part of Everyday Conversations
You don’t need a formal AI task force to get started: A small step can be just as effective. During a team meeting, take a few minutes to discuss a recurring challenge and ask whether AI might help. If someone has tried a tool, invite them to share what worked and what didn’t.
These conversations help normalize learning and create opportunities for employees to discover practical uses for AI in their own work. Over time, they can also help supervisors better understand where AI is creating value and where it’s creating frustration.
The Bottom Line
The most effective supervisors won’t be the ones with the deepest technical expertise. They’ll be the ones who create an environment where employees feel comfortable asking questions, testing ideas, and learning together. When conversations focus on solving problems, improving service, and supporting employees, AI becomes less intimidating and a lot more useful. And that’s often the best place to start.
Looking for more tips on becoming a better supervisor? Join us next month on Monday, Dec. 14 at 4-4:30 p.m. ET/1-1:30 p.m. PT for “The Emotionally Intelligent Supervisor.”



Leave a Reply
You must be logged in to post a comment.