AI Is Not Here to Replace Us. It Is a Tool in Our Tool Belt.

One of the things that every AI conversation inside a government team has to settle is whether the tool is going to take the job.

It is not. That is the wrong fear and the wrong frame. AI is a new entry in your tool belt. The job is still the job. The communicator is still going to write the press release. The analyst is still going to read the 311 data. The communications director is still going to set the strategy. AI is a way to do those things faster, with more leverage, and with more time left over for the parts of the work that require a human in the chair.

The teams that get this reframe right early are the teams that move past the existential AI debate and into the practical one. The teams that do not get stuck arguing about whether to adopt at all, which is a debate the rest of the workforce has already had.

Key insight: AI is not the new job. It is the new tool. The job did not change. The tool belt got bigger.

When I was in the TV business, I was originally given a crew to shoot so I could focus on writing and producing.

Then the industry changed. The demands on me and my skills changed too. I needed to shoot. I needed to edit. I did not have the budget to hire huge teams, so I needed to figure it out on my own. The job description on paper did not change. The tools I was expected to hold changed completely.

That is the moment government is in with AI right now. The job is the same. The tools are different. The people who adapt are the people who get curious about the new tools when they show up, not the ones who wait until the description on paper changes.

This is the same instinct that made me pick up a camera then is the one that will make a 311 supervisor pick up an AI tool now. Curiosity, permission to figure it out, and the time to try.

Key insight: The job did not change for me at MTV. The tools did. And the people who picked up the new ones kept working.

Just like staff go to annual trainings or conferences, AI learning has to be on the calendar.

A recurring time where anyone curious about how to use AI at work can show up, ask the question they were embarrassed to ask in a one-on-one, and leave with one thing to try. Open to anyone in the org. Open to the person whose title does not include "digital" or "data" or "IT." That is the open invitation that is actually open.

The other piece is permission to learn outside the building. Send people to conferences that are not government-focused. Let them watch what the private-sector communicators and the agencies serving other industries are doing with AI, and bring those ideas back into the room. Government is rarely at the edge of practice with new tools. The people doing the work do not have to stay there.

The people who feel like they need permission to learn are the ones whose growth is most expensive when it does not happen. The people who already feel permitted are out there learning anyway. Closing that gap is a calendar move, not a budget move.

Key insight: AI learning is a ritual on the calendar, not an event that happens once. And the invitation needs to be wider than the IT org chart.

The cost of staying still is not evenly distributed.

In a March 2026 Lean In / Wired Research survey of 1,000 US adults, women were less likely than men to be encouraged by their managers to use AI at work, 30% versus 37%. The encouragement gap sounds soft, but it is the most measurable form of the permission problem. Where managers are quietly encouraging some team members to try AI and not others, the team that is supposed to learn together does not learn together. (Lean In, March 2026)

The gap closes where the culture is intentional. Where the open invitation is actually open. Where the conference budget reaches the comms team, not just the IT team. Where the recurring AI time on the calendar is not the early-adopters club.

The same human-first principle I have been writing about for how cities use AI with residents applies to how we use it inside our own walls. AI is not Big Brother. It is a tool. And the tool has to land in the hands of everyone on the team, not just the ones whose manager remembered to mention it.

Kristy Dalton has spent her career building the field of government social media. When we talked on the podcast, she named something I keep thinking about. The thing that keeps people from trying AI at work is not always missing skills or missing time. It is the fear that learning openly will look like cheating.

"There's also a group of people who also are hesitant about AI because of the fact that they feel like it's cheating because you're using ai. When government communicators come up through the ranks and learn how to write and to tell stories and all of that and this is cheating if you're just gonna leverage that and post it on social."

Kristy Dalton, Episode 11

That fear gets quieter inside a culture of learning together. The faster the team builds the rituals, the faster the cheating reflex stops doing quiet work in the background of every meeting where AI comes up.

Who on your team has been waiting for permission to learn AI? Reply and tell me. I read every reply.

Talk soon,

Dana

Was this forwarded to you? Subscribe to CityCX here.

Know a manager, CDO, or comms director whose team has been waiting for permission to learn AI? Forward this email.