Before Your Fleet Uses AI, Build Trust First

 I don't think AI is going away.

If anything, it's becoming another tool fleet managers will use every day.

Some of us are already using it to summarize reports, organize meeting notes, draft policies, or help pull together presentations. Fleet management software is starting to include AI, too.

That doesn't surprise me.

What does surprise me is how quickly some organizations are adopting it without first deciding what role they want AI to play.

We've seen this before.

Every time a new technology comes along, there's a rush to figure out what it can do. The better conversation is figuring out how it fits into the way your organization already operates.

AI shouldn't be any different.

Trust Starts Before Anyone Opens the Tool

Recently, RTA surveyed fleet professionals about AI.

I expected cybersecurity and data privacy to be the biggest concerns.

Instead, the top answer was trust.

Can I rely on what AI gives me?

I think that's exactly the right question.

Not because AI isn't useful. I use it myself.

But because public fleet leaders don't have the luxury of assuming the answer is right.

When you're responsible for taxpayer dollars, procurement decisions, employee safety, and regulatory compliance, "the AI said so" isn't much of a defense.

If someone asks how you arrived at a recommendation, you still need to be able to explain it.

AI Doesn't Change Who's Responsible

One of the mistakes I see is treating AI like it's making the decision.

It isn't.

It's making a recommendation.

There's a big difference.

If AI helps you organize information before a budget presentation, great.

If it gives you a starting point for a replacement policy, even better.

But before that document goes to your finance director, your city manager, or your governing board, your name is still the one attached to it.

Would you be comfortable defending that recommendation if someone asked where it came from?

That's the part I don't think changes.

Technology keeps evolving.

Responsibility doesn't.

Fast Answers Still Need Experienced Eyes

One thing I've noticed about AI is that it's incredibly confident.

Sometimes it's confidently right.

Sometimes it's confidently wrong.

The challenge is that it usually sounds exactly the same.

That's why I never assume the first answer is the final answer.

If AI summarizes a regulation, I go read the regulation.

If it compares maintenance strategies, I compare those ideas with what I know about the fleet.

If it drafts a policy, I rewrite it until it reflects how the organization actually operates.

That's not because AI failed.

It's because reviewing information has always been part of our job.

Give People Expectations Before They Create Their Own

One survey result really stood out to me.

Three out of four respondents said their organization has little or no formal guidance around AI.

Meanwhile, people are already using it.

That's where things can get inconsistent.

One person uploads information they shouldn't.

Someone else uses a tool that hasn't been approved.

Another employee assumes AI's answer doesn't need to be verified because it "looked right."

None of those situations start with bad intentions.

They start with unclear expectations.

That's why governance matters.

Not to slow people down.

To make sure everyone is working from the same playbook.

A Simple Way to Keep Yourself Grounded

During a recent conversation on The Fleet Success Show, Jen Brown shared the framework her team uses whenever AI is involved.

Verify. Can you confirm the information?

Origin. Where did it come from?

Comply. Does it align with your organization's policies and public-sector requirements?

Authority. Who's making the final decision?

That last question is my favorite.

Because if we ever get to the point where the answer is "the AI," we've forgotten what leadership looks like.

The Best Use of AI Isn't Replacing Work

One thing I hope doesn't happen is that we start measuring AI by how much work it replaces.

I'd rather measure it by how much time it gives back.

If AI drafts the first version of a report, that's more time talking with technicians.

If it summarizes a meeting, that's more time walking the shop.

If it organizes information before a budget meeting, that's more time thinking through the questions you'll probably be asked.

That's where fleet managers add value.

Not typing.

Thinking.

Judging.

Explaining.

Making decisions.

Good Technology Still Needs Good Leadership

Fleet has adapted to change before.

We learned new fleet management systems.

We adopted telematics.

We figured out mobile inspections.

We'll figure out AI, too.

The technology isn't what concerns me.

Making sure we use it wisely does.

Because at the end of the day, AI can recommend.

It can summarize.

It can organize.

But it can't stand in front of your city council, defend your replacement plan, explain your procurement decision, or answer questions during an audit.

That's still our job.

And I think it always will be.

Frequently Asked Questions

Should public fleet departments use AI?

Yes. AI can help summarize information, draft communications, organize data, and support analysis. Like any tool, it works best when fleet leaders verify the output and remain responsible for every decision.

What are the biggest AI risks for fleet organizations?

One of the biggest risks is relying on inaccurate information without verifying it. Clear expectations, approved tools, and thoughtful review processes help fleets use AI responsibly while reducing unnecessary risk.

Does my fleet need an AI policy?

Absolutely. AI is already finding its way into day-to-day work. Establishing expectations around approved tools, acceptable data, verification, and accountability helps everyone use it consistently and responsibly.

This article was inspired by a recent episode of our podcast. Check out the full episode for even more tips and tricks: