Webinar Recap: Stop Chasing AI Use Cases. Start Solving Problems

We recently held a webinar with our team about where AI is actually landing in local government and why most municipalities are approaching it in a way that's going to cost them time and money. The conversation surfaced three things worth your attention, depending on where you sit right now.

If you haven't started yet

You're probably drowning in AI hype. Every vendor has it. Every news story mentions it. Council is asking about it. So the instinct is to ask: where should we deploy AI?

That's backwards. Natasha Singh, one of our senior consultants, was direct about this: "AI has to earn its place by solving the real problem. We cannot automatically assume AI is the solution for everything." The better starting point is a much simpler question. What is the problem? What is the opportunity? Where are residents struggling? Where is staff doing repetitive work? Where do applications come back incomplete? Where is information scattered?

Once you understand the real pain, then you ask whether AI is part of the answer. Sometimes it's not. Sometimes the answer is staff training. Sometimes it's a simpler form or better web content. "AI is one possible response and not the default answer," Natasha said. That distinction changes everything because you'll waste time and money if you're hunting for places to apply AI instead of solving actual problems.

We published a Responsible AI Framework this year to help with exactly this. The framework covers who owns what (council, CAO, senior management), what principles guide your work, what you absolutely cannot do, and how to stage your adoption from explore to norm to scale. If you're at this stage, grab the framework and spend an hour on the governance section and the red lines. Then pick one small problem, something real that's costing you time or quality, and assemble a cross-functional team around solving it.

If you're already experimenting

You're probably running pilots. Maybe several. The risk here is that you've started adopting AI without first asking whether it's solving the problem or just becoming a trend.

Ben Perry put it plainly: "We think you should experiment. We think you should learn. We think you should adopt and we think you should adapt." But the experiment only works if you measure the outcome. Are you actually reducing the number of incomplete applications? Are you actually freeing up staff time for complex work? Are you actually improving the service? Or are you just running an AI chatbot because everyone else is?

This is where governance kicks in again. You need to know what you're measuring before you start. You need accountability for the results. You need to be honest about what's working and what's not. The other thing worth flagging here is that you should make sure you're not confusing automation with AI. Natalia Madden raised this directly: "There's automation and then there is AI. And we really need to think clearly about those two topics." Sometimes you don't need AI. Sometimes a simpler form or a better workflow does the job. The risk is that you over-engineer the solution because you're excited about AI.

If you're doing any of this

Governance is required. It's not something you hand off to IT and forget about.

Ben was clear on the red lines: "Under no circumstances should AI fully author a report without being reviewed, without having the human in the loop." You own the outputs. You own the decision to deploy. You own what happens if something goes wrong. That means you need governance in place before you scale. You need accountability. You need people trained on what they can and cannot do. You need documentation of your process.

This is about being responsible. Bill 194 is coming to Ontario municipalities, probably in 2027 and beyond, and it will require this stuff anyway. Get ahead of it now. The other piece that matters is that you should disclose when you're using AI. Let people know. The narrative has to shift from "we're using AI secretly" to "we're using AI safely and we're measuring the impact."

You're in the first inning

Ben's framing on this is worth holding onto: "We're in the bottom of the first inning of this game. There's a long way to go in this game." This is an opportunity that requires thoughtfulness, not speed. Most municipalities are not at the bleeding edge and that's fine. You've got time to learn, to experiment, to build governance, to measure what works. The cost of entry right now is curiosity and a bit of your time. That's the lowest it will ever be.

But the cost of waiting is your competitors moving faster, your staff falling further behind, and the gap widening. So start small. Start with a real problem. Build the governance. Measure the outcome. Learn from it. Then expand.

We've published the Responsible AI Framework on our website. Download it. If you want to talk through where you are and where you want to go, whether you're exploring, experimenting, or scaling, book a call with us. We'll help you sort out what's real and what's hype.