Manufacturers Are Using AI. Most of Them Are Barely Scratching the Surface.
I came across a piece from IT Wire about where manufacturing IT stands right now with automation and AI adoption. The numbers in the article are all over the place, but one number jumped out at me and I could not let it go.
On average, a manufacturer does 3.08 prompts a day to an AI tool. Across all industries, that average is 3.28. In tech, it is 4.72. So manufacturers are running almost a full prompt a day behind everybody else.
Why That Number Bothers Me
A prompt is just a question to an AI. And if you are treating it like a search engine where you type in a question and read the answer, you are not really using the tool. That is not the point.
The way I use AI, I am probably sending hundreds of prompts a day. Not because I am asking the same question over and over, but because I am constantly building things. Automations. Reports. Capabilities I do not want to handle manually. I am giving it commands and making it do work, not just asking it to summarize something.
So when I see 3.08 prompts a day, what I read is: most manufacturers do not know how to use this yet. Take that for what you will.
The ROI Problem Is Real, But It Is a Usage Problem First
The article also makes the point that manufacturers are ahead on proactive automation compared to other sectors, which is good news. But they still cannot clearly prove what AI is returning for them.
I think those two things are connected. If you are only doing three prompts a day, you are not giving the tool enough work to show you anything meaningful. You are not going to see ROI from occasional use. You see it when the tool is doing real, repeatable work for you every single day.
If you are a manufacturer and you have already started experimenting with AI tools, the question worth asking is not just whether you have the tool. It is whether your team is actually using it to do things, or just poking at it once in a while.
What I Would Do If This Were My Shop
- Stop counting prompts and start counting tasks. What is the AI actually completing for you each week?
- Pick one thing you or your team does manually and repeatedly, and make that the AI’s job. Not a one-time question, a standing task.
- Once it is doing that task, measure the time saved. That is your ROI data. It does not have to be complicated.
- Then build from there. One task becomes two. Two becomes ten. That is when the numbers start to look interesting.
If you want to talk through where AI could actually fit in your operation, give me a call.
Questions about how this affects your setup? Schedule a free IT consultation and we’ll walk through it with you.
Wherever your team is based, ACS provides hands-on managed IT support — see how we help businesses in Chicagoland.