AI gets talked about like it’s already running the factory floor. For most small and mid-sized manufacturers, that’s not reality. The promise is big—autonomous systems, seamless digital operations—but the day-to-day still looks a lot like it always has.
What’s changed is the pressure. Fewer skilled workers. Material costs that won’t sit still. Customers expecting faster turnarounds and instant quotes. AI isn’t stepping in as a silver bullet—it’s starting to show up as something more practical.
There’s no shortage of optimism. Most manufacturers believe AI can help. But many aren’t seeing results yet. That disconnect usually comes down to tools that weren’t built with manufacturing in mind. Generic solutions don’t understand the nuance of job shops, custom work, or the way production actually flows.
The shift now is toward something more grounded. Instead of chasing transformation, manufacturers are looking for ways to make the work they already do a little easier.
That’s where AI starts to make sense.
Not as a standalone system. Not as a complete overhaul. More like a quiet addition to the tools already in place.
Think about quoting. Instead of starting from scratch every time, AI can pull from past jobs and build a first draft in seconds. Estimators still make the call—but they’re not buried in repetitive work.
Or take Bills of Materials. Pulling data from drawings, building out BOMs, catching inconsistencies—those are the kinds of tasks that eat up time and introduce errors. Automating that process doesn’t replace expertise. It just clears space for it.
Scheduling is another pressure point. Shops run on tight timelines, and small disruptions can ripple quickly. With the right data, AI can flag potential bottlenecks before they slow everything down.
None of this is flashy. That’s the point.
The manufacturers seeing progress aren’t trying to do everything at once. They’re picking one problem—usually something repetitive or time-consuming—and starting there. Layer by layer.
And the most effective approach keeps people in the loop. AI handles the data-heavy lifting. The team stays in control of decisions that actually move the business forward.
Where AI lives matters, too. When it’s built into existing systems—especially ERP platforms—it has the context to be useful. Without that, it’s just another tool sitting on the sidelines.
There’s still a lot to figure out. But the direction is getting clearer. Less hype. More application. Less about transformation, more about traction.
For manufacturers trying to make sense of it all, the takeaway is simple: start where it hurts. Solve one problem well. Then build from there.
Read more here.
Polaris MEP’s own AI Mini Survey, is in agreement with the findings found in the above linked article and in turn is actively providing services to manufacturers that meet their needs, such as our recent Vertikal 6 webinar.