_{Don’t stop believin’ in the Journey}

Here's the very real future: you walk into your office, open a chat, and ask "what is my most dangerous bus route?" And AI looks across your CAD/AVL data, your customer complaints, your incident reports, maybe even your socioeconomic overlays… and gives you an answer in minutes. Not after a six-month study or a $200k RFP cycle. Minutes.

That's not science fiction. That's Phase 5 of the AI Journey: AI-Integrated. Every decision you make will be informed by connected data, every analysis that used to take a full quarter can happen before lunch. The agencies that get there first won't just be more efficient, they'll be operating in a completely different reality than the ones still waiting for their annual consultant report.

So naturally, every transit director asks me the same thing: "How do we get there?"

There's an order of operations, and skipping steps is the great risk of all.

The five phases divide into two halves, and the line between them is everything. Phases 1 through 3 are your individual leadership journey. Phases 4 and 5 are your organizational leadership journey. And you cannot — I really mean this — you cannot get to the organizational side until enough people have walked the individual side first.

This is where it all starts. Phase 1 is awareness — you've tried ChatGPT, maybe typed something carefully, got a decent-ish answer, thought "huh, that's kinda cool." Phase 2 is literacy — you've upgraded to paid thinking tools, and AI stops being a novelty and starts being a genuine thinking partner. Phase 3 is leadership — you've developed taste across multiple tools, you've connected AI to your context, and you're completely reimagining how your actual work gets done.

I saw this play out beautifully a few weeks ago in Colorado. I spent two days with a group of transit leaders from across a mountain region, each with completely different problems. One person needed to automate their annual NTD reporting, which normally takes three to four weeks. Another was navigating a political debate about rail versus bus and needed evidence without spending hundreds of thousands on consultants. Someone else was developing a regional coordination strategy across multiple agencies. Another was rethinking how their organization develops semiannual operator paddles.

Every single one of them was on their own individual journey. Different starting points, different problems, different comfort levels. But by the end of those two days, each had built something real — not because I told them what to build, but because they explored what was possible for their specific situation. One person went home and built a complete NTD reporting tool over the weekend. On their own. That's what happens when someone walks through Phases 1, 2, and 3 at their own pace with the freedom to explore.

Here's what happens when agencies try to skip to Phase 4 without the individual journeys underneath: they centralize around the wrong tools. They build policies that restrict instead of empower. They spend six months evaluating platforms when their people haven't even learned to talk to AI yet. An enterprise adoption will fail if  nobody has the individual fluency to make it work.

Phase 4 (AI-Enterprise) — is when your organization picks its centralized tools and starts building agentic workflows. Phase 5 (AI-Integrated) is that connected-data future where decisions happen in minutes. Both are real, and both are coming. But they only work when they're pulled by a critical mass of people who've already walked Phases 1 through 3, not pushed from the top onto people who haven't.

The bridge between the two halves? Give your people a green light with two guardrails. Build your AI Task Force of cross-functional champions leading by example. Let the individual journeys compound until enterprise adoption becomes the natural next step — not a mandate, but a pull from the people who are already there and ready. It’s the electric current of change.

That Phase 5 future is real. But it starts with you opening a tool, being messy, and discovering what's possible for your work first. The technology is actually the easy part. The individual journeys are the key.

TransitGPT is where I share practical AI insights for transit leaders. Subscribe if this resonates or forward to a colleague who needs it. And hit reply — I read everything.

Go shatter your glass ceiling.