Most city and town offices run on a handful of people doing too many things. The website. The council packet. The meeting minutes. The public notice that has to be accurate, plain enough for every resident to follow, and posted on time. None of it is hard, exactly. It is just slow, and it never stops.
That is the work AI is actually good at. Not replacing your staff's judgment. Taking the repeat work that sits between them and the judgment.
Here is what that looks like in practice, using tools I built for my own work.
One update, posted everywhere it needs to go. A road closure or a schedule change has to land on the website, the newsletter, and social, worded consistently each time. Rewriting the same notice three ways by hand quietly eats an afternoon. I built a small system on top of Claude that holds my voice and rules in one place and drafts each version from a single update. I still edit. But I start from drafts, not a blank page.
A repeatable pipeline instead of a one-off scramble. For one of my publishing projects, the steps were always the same: research, draft, check the facts, make an image, publish. I turned that into a fixed pipeline the AI walks through every time. The work does not disappear. It happens the same way every time, so nothing gets skipped and the quality does not depend on how busy the office is that week.
The boring tasks, done reliably. Formatting a council packet. Drafting first-pass meeting minutes from a recording. Checking a PDF against accessibility rules before it goes live. People put these off, and AI handles them well, because the standard does not change.
Where this fits a city or town
- Meeting minutes and agendas. Turn a recording or rough notes into a clean first draft of the minutes. Staff reviews and approves instead of typing from scratch.
- Plain-language public notices. Take a dense ordinance or notice and produce a version residents can actually read, plus a translated version for the languages your community speaks.
- Resident questions. Draft consistent answers to the questions you get every week (trash days, permits, payments) so nobody reinvents the reply each time.
- Accessibility checks. Flag PDFs and pages that fail ADA and WCAG rules before they are posted, instead of finding out after a complaint.
- Grant and report drafting. Pull your existing records into a first draft of the report a grant requires, in the format it requires.
The pattern is the same every time, and it is the part worth keeping:
- Find the work staff does over and over.
- Write down how it should be done, once.
- Let the AI handle the repeat passes, and keep staff judgment for the decisions that actually need a person.
You do not need a custom platform or a data team to start. You need one repetitive task and a clear definition of what "good" looks like. That is usually enough to get the first few hours back.
At NavigateTomorrow, this is how we think about AI for the cities and towns we work with. Not a product to buy. A way to make the work your office already does more reliable and less manual, so a small staff spends its time serving residents instead of retyping documents. We bake AI into every website we build, so even there you get the advantage.
Want to talk through where this fits for your town?