“Humans will then become almost like a kind of team leader”

Editorial team: Dr. Zein, let's start with a broader perspective: when you look at the entire field of engineering, where does AI stand today? Where do you see the greatest maturity, and where are the biggest gaps?

Dr. Axel Zein: In general, we are still at the very beginning of AI in engineering - really at the very beginning. Compared to other areas, the maturity level is very low. In my observation, AI is most advanced in software development. This is probably because software developers build the technologies themselves and have a natural affinity for them.

In traditional engineering disciplines like mechanics, electrical engineering, or PCB design, AI is almost non-existent. In Europe, we generally find it more difficult. In electrical engineering, we at WSCAD are currently the only ones who already have AI functions in productive use. This also shows that there is a lot of talk about AI, but practically little is happening so far. But it will come, for sure.

Editorial team: Many companies don't fail due to lack of goodwill; many start pilot projects. But then scaling doesn't succeed. What are the reasons from your experience?

Zein: The most important question at the beginning of any AI project is not: "Which technology are we using?" but: "What problem do we want to solve?"

We made this mistake ourselves at first because we approached it from the wrong side: What can we do with AI? That's nice, but pointless. You need use cases that bring real benefits. For that, you have to bring together data, developer know-how, and domain expertise cleanly. And you need people who are willing to think outside the box. This working out of possible use cases is fundamental - before you even think about tools or technology. And this phase is greatly underestimated.

Editorial team: That means mindset plays an enormous role?

Zein: Absolutely. No AI project works without an open mindset. But it's not enough to just be open - you also have to want to work differently.

A good example from recent history is the introduction of robots in manufacturing: at first, everyone was afraid for their jobs. In the end, it was found that automation allowed for cheaper production - and even more jobs were created as a result. Not the same ones, but new ones. That's exactly how it will be with AI.

Editorial team: Will this change the education and role of engineers?

Zein: Yes, but not in the sense that professions will disappear. Engineers need to learn to work with new tools. With AI, engineering is already significantly faster in some areas today - sometimes the user only needs a fraction of the time. This means: you can handle more projects in the same amount of time.

Just like in manufacturing back then: the employee no longer welded themselves but learned to set up and monitor the robot. In engineering, it's similar: you work differently, but not less.

Editorial team: Many people are afraid - both of losing their jobs and that AI results might be faulty. How do you build trust?

Zein: By openly addressing fears, not sweeping them under the rug. When the topic is brought up, historical parallels often help - for example, from the horse-drawn carriage to the train: yes, the coachmen were gone. But every train created far more jobs than a horse-drawn carriage ever had.

When it comes to reliability, it's crucial: you should never work purely with AI models in engineering. AI can hallucinate. That's why we combine strict rule-based systems ("if-then rules") with AI models. This produces robust results - and thus trust. Only in this way can you achieve acceptance.

Editorial team: Today we see AI functions in CAD systems that facilitate individual tasks. But if AI agents take over more and more in the future - will CAD become an interchangeable commodity?

Zein: What we see today is just the beginning: AI features that make existing CAD tools smarter. That's useful, but not yet transformative. The real leap comes when AI truly takes over parts of engineering. We've made a small first step with our automatic control cabinet generation - but that just scratches the surface.

Our vision is a system that acts like an experienced electrical designer. You enter requirements - not fully naturally - and the system asks follow-up questions, suggests solutions, and designs. The human then becomes almost like a team leader, guiding a virtual designer.

With our current solution, some customers are already achieving 50% time savings today. Imagine what happens when AI takes over large parts of engineering - that's several orders of magnitude more.

Editorial: When could something like this become a reality - 2027, 2028?

Zein: You might see initial developments in 2027. Agents can already perform simple tasks today, but engineering is extremely complex: strict rules on one side, creative conclusions on the other. This balance is not trivial. How quickly or slowly AI-based solutions will actually emerge here, no one can say today. Especially since we don't know what AI technologies will be available in 2027. I wouldn't have thought twelve months ago that we would be at this point today - and I expect more surprises.

“AI is change management - at the speed of light”

Artificial intelligence is long more than a buzzword in the industrial environment. At the Automation NEXT Conference 2025, Dr. Axel Zein, CEO of WSCAD GmbH, demonstrated from a manufacturer's perspective how engineering processes can not only be optimized but fundamentally changed by AI.

Editorial team: Who actually has the advantage in the race for AI engineering? USA? China? Europe?

Zein: When we look at AI globally, the ranking is clear:

1. USA

2. China

... nothing for a long time ...

Then maybe the UK or India. Unfortunately, Europe doesn't play a big role because Europe regulates - and watches. That's unfortunately the reality. There are exceptions like Mistral in France, but they are too small in a global comparison.

Our advantage at WSCAD is that we operate in a very small niche. Big players like Google or Autodesk could technically do what we do without any problem, we have to be realistic - but for them, our market is far too small.

This also explains why Autodesk has hardly developed its product "AutoCAD Electrical" for years: It is weak compared to modern electrical CAD systems, but for Autodesk, this area hardly plays an economic role.

Editorial team: Then let's zoom out of your niche again: Where do you see the biggest low hanging fruits for AI in engineering?

Zein: Due to the maturity level described at the beginning: In all of engineering. In mechanics, electrical engineering, PCB design - everywhere. In fact, almost no AI is used in engineering today. If someone uses AI, it's more likely GPT tools for writing requirements - but not in actual engineering. We are just at the beginning here.

Editorial team: Finally, let's assume a manager reads this interview and wants to start with AI this year. What would be the first steps?

Zein: First: Form a small coalition of the willing - people from the specialist area who are curious and eager to rethink things. Add one or two software developers or data analysts.

Second: Give the task of finding use cases. These cases should then be evaluated based on potential impact and effort.

And third: Start small, get going immediately, report weekly. Don't start a six-month project right away. A small project should ideally show visible success quickly - so that others in the company say, "That's exciting, I want that too."

And very important: The boss must be on board. Either it is important, then the leadership level must pay attention. Or it is not important - then maybe it should just be left alone.

AI in engineering at the packaging machinery conference

We will also explore the question of how AI is currently and will in the future impact engineering at the 3rd Packaging Machinery Conference, which will take place on June 16 and 17, 2026, in Nuremberg.

Our speaker on this topic is Dr. Daniel Eckertz, group leader innovation engineering at Fraunhofer IEM, with whom we recently conducted an interview.

For all further information on the platform of packaging machinery, please visit https://www.packaging-machinery-conference.de/