I’ve been thinking about how editorial processes have changed in the publishing world. Before printing an issue, we would regularly send a PDF of the issue to an outside copy editor, who would notate grammatical errors and highlight areas that needed clarity. This process no longer happens.

While the shift away from outside copy editors started before the AI revolution, copy editors are not a growth position, because AI can do a lot of the job. (But not all of it.)

Today’s business mandate is “Do more with less and do it faster.” AI is accelerating this mandate at warp speed. Like publishing, this has me thinking about how fleets would expand, contract, or shift, and how the fleet workforce must adjust to these changes. 

More Efficient Operations, Fewer Vehicles

The prime directive of any fleet manager is to increase the efficiency and utilization of every vehicle in the fleet, and many AI tools already promote doing more with less. 

AI-enabled routing and dispatch are cutting idle time, optimizing routing, and helping to avoid deadhead runs. Similarly, predictive maintenance extends vehicle life and reduces downtime, which extends lifecycles and reduces replacement frequency. 

AI can also analyze utilization tracking data to help downsize or consolidate fleets without sacrificing performance.

With these tools, the work that required 125 vehicles can now be done by 100. To be clear, limiting fleet growth in these ways is a good thing. 

A Shift in Onsite Jobs

Outside of an organization’s internal fleet count, broader business dynamics affected by AI could lessen or expand the need for fleet vehicles. 

B2B sectors that once required in-person reps in sales fleets may transition to AI-supported digital platforms. Remote diagnostics and monitoring tools to service machines and diagnose patients are already being used to replace physical site visits. On the flip side, AI will likely enable new services, like home healthcare and mobile vehicle repair. 

And think about the gargantuan power needs of AI:

Hyper-scale data centers require infrastructure and the power of a small city. Consider the electricians, telecoms, renewable energy providers, and cloud computing companies that will use fleet vehicles to construct and service them. The investment capital is already in the trillions.

The Autonomy Factor

In the long term, the shift to autonomous vehicles, dependent on AI, will reduce the need for company-owned fleets. Shared autonomous networks could replace traditional fleet logistics. This would happen in urban delivery and ride-hailing before replacing traditional corporate fleet use. I’ll be retired by then (I think). 

But even in a world of robotaxis and autonomous delivery, fleets don’t disappear; they evolve. Autonomous operations still require dispatch and routing centers, scheduled maintenance and cleaning, recharging hubs, and real-time tracking. 

In fact, vehicle usage in driverless fleets could increase with continuous operation, requiring more robust back-end fleet management tools and staffing.

Here’s the Double-Edged Sword 

Yes, AI can help you do more with less, but that will open the door to staff reductions. Managing 1,000+ vehicles with a smaller team will likely become standard.

If something goes wrong — an error in the maintenance model, a compliance misfire — it’s the human, not the AI, who has to clean up the mess. If your team is stretched thin, those mistakes are harder to catch and more costly to fix. 

In other words, AI could expand a fleet manager’s reach but also narrow the margin for error and create new dependencies on tools that are only as good as the humans managing them. Just ask Air Canada. No, the chatbot isn’t liable — you are. 

Back to my copy editor analogy: While Grammarly (which I’m using as I write this) is great for proofreading and even for improving clarity and readability, it can’t replace the other part of copy editing that is most valued — questioning me for context or if something doesn’t sound right. 

I need to actively create other workarounds to ensure accuracy and clarity in my work that don’t add extra expense, such as a copy editor. Otherwise, I’m failing a quality check that could get me in trouble. 

I miss the human copy editor, not because they caught typos, but because they asked the right questions. That’s where AI still falls short. 

In fleet, we’re entering a new world of “exception reporting,” where the value shifts to those who can spot subtle problems that technology cannot. 

Efficiency is essential, but hopefully not in deference to judgment and accountability. You can’t outsource responsibility to a tool. We just need to make sure our organizations understand this fully and value it.