Sathish:
“There is massive investment happening in manufacturing AI right now, but many initiatives fail. From your experience, what is really going wrong?”
Mike:
“The topic deserves nuance. You see headlines saying 50% or 80% of AI projects fail in manufacturing. You have to ask who is publishing those headlines and why, and whether the premise is actually accurate. There is a lot of AI that has been sliding under the radar for the last 10 years, successfully incorporated into organizations but no longer recognized as AI. Chatbots that answer customer calls? That is AI. We just call it the phone system. Facial recognition used every day across facilities? That is AI. We call it security. Inspection cameras with embedded algorithms detecting defects in real time? That is AI. We call it quality control. As soon as AI becomes successful, it gets renamed into something else and loses credit for being AI.”
Sathish:
“So if it is not the technology itself, where do things go wrong?”
Mike:
“Two friction points. The first is choosing the tool before choosing the business case. AI has created enormous pressure on CEOs and boards to announce that they are doing something with AI. There is pressure to act. But there is a real disconnect between having an AI tool and using it to perform an existing business function. There has to be recognition upfront of what problem you are solving: cost savings, quality increases, better customer satisfaction. That clarity is lacking in a lot of big AI projects. The second issue is the skills gap. Even when leadership formulates a solid plan, who implements it? Education and training institutions are struggling to produce graduates who can take AI from theory into practice.”
Sathish:
“How big of a challenge is data quality and infrastructure?”
Mike:
“We are unprepared on the people side, but we are also unprepared on the underlying data structure of the plant. There is a lot of manufacturing running production equipment as islands of automation without much connectivity between them. Different names, different tags used to describe the same data. Just trying to throw all of that into some AI program and have it make sense of it is probably not going to work. Factories are throwing out terabytes of data every day, but is it tagged properly? Can a human read it and understand it? There is a concept called the Unified Namespace which is an attempt to standardize data so it almost self-explains itself: which plant, which city, which piece of equipment, which sensor. Once you have that, then tools like ChatGPT and Claude might be able to do remarkable things. But the old principle still applies: garbage in, garbage out.”
Sathish:
“What do manufacturers misunderstand about ROI when investing in AI?”
Mike:
“It comes back to the ROI not being clearly defined before a project starts. If you do not know where you are going, how do you know if you are lost? The goal can be modest. A 5% or 10% reduction in waste or time would be a very nice win. But you should be able to measure it. Two areas consistently show strong ROI: quality control, especially in sectors like semiconductors where people are looking through microscopes to check solder joints, and predictive maintenance, where you are monitoring equipment health to eliminate unplanned downtime. Festo AX monitors something as simple as how fast a cylinder piston extends and retracts, and based on timing and acceleration, can warn you that a component is at 85% of its lifespan and should be swapped in two weeks. Most people would never think of that as AI. But it is exactly the kind of targeted application that delivers real ROI.”
Sathish:
“What is the biggest mistake leadership teams make with AI adoption?”
Mike:
“You have to address the elephant in the room: fear. People think they are going to implement the system and then lose their job. You have to provide emotional and psychological safety first. It is like Maslow’s hierarchy of needs. At the very bottom is safety. Nothing else happens unless people feel secure. If you are implementing a million-dollar AI system, you should have about a million dollars also set aside for workforce development and organizational change. If you just buy the cool shiny technology and do not invest in people, that is where failure happens.”
Sathish:
“Should manufacturers start with pilot programs or go straight to full-scale implementation?”
Mike:
“Pilot programs are probably the easiest way to go. But there is a concept called pilot purgatory: you run a pilot and it never becomes more than just a pilot. Before the pilot starts, define in writing what success looks like. We are going to install a vibration-sensing AI system for predictive maintenance on this line. We expect 20% fewer unplanned stoppages. Then the pilot either hits the target or it does not. If it works, you have a clear case for scaling it to every similar machine. If it partially works, say 12% instead of 20%, you can evaluate whether that ROI is still worthwhile. But if you do not define success criteria upfront, the pilot can technically succeed and nobody knows.”
Sathish:
“What skills will be in demand among manufacturing professionals?”
Mike:
“Soft skills are making a huge reemergence. We are taking the job of being a robot out of a person’s job. For a long time, we hired people as living robots: pick up bricks from here, put them there all day. When you remove that, what is left? Teamwork, communication, cultural awareness, decision-making at every level, leadership from the floor. The CNC operator used to place metal into a machine, close the door, and press a button. Now robots feed the machine. The machine turns itself on. So maybe the role becomes a production specialist whose job is to ensure all machinery is performing the right quality and quantity of work on their shift. That means reading screens, interpreting sensor data, watching for warning lights, and making judgment calls. These jobs require the next level up in thinking, which most people are capable of. It was just never required of them before.”
