Artificial intelligence in product development: hype, helpful tool or lever for change?

Last December, I attended a meeting of the professional advisory committee for Industrial Product Design in Groningen. We discussed current developments in our field, with one clear main theme: artificial intelligence (AI). Not as an abstract vision of the future, but as a concrete question: what does AI really mean for product development?

By 2025, generative AI had firmly entered the daily work of many professionals. Chatbots, image generation and code assistance have quickly evolved from curiosities into everyday tools. In sectors such as education, software development, marketing and consultancy, the impact is clear. But when the question came up what this means for physical product development, no one had an immediate, concrete answer.

Within PEZY, we were exploring those same questions at the time. Which AI tools truly add value to our work? How do we safeguard quality and the confidentiality of information? And perhaps the most important question of all: where does the human hand remain essential?

I have been following the development of AI for years and have experimented with it myself in various stages. From early tools such as DeepArt and Artbreeder to running Stable Diffusion on my own laptop. At the same time, I have always remained critical. Where does AI truly add value? Where is it mainly hype? And where does the risk arise that speed is mistaken for quality?

That question became even more tangible for me when one of the people I was speaking with mentioned a LinkedIn post by Guido Stompff, professor of Design Thinking at Inholland University of Applied Sciences. He described how his annual “coffee filter challenge” had failed for the first time this year. For years, students had come up with a wide variety of ideas. This time, almost every group ended up with the same idea. The reason? Chatbots. What used to be a creative exercise with fifteen minutes of lively interaction became a three minute exercise between students and their phones. In silence.

It is a striking example of the paradox of AI: more speed, but less variety, less engagement and sometimes less genuine innovation.

Productivity, with caveats

Recent research shows a similar picture. AI can deliver productivity gains, especially in software development. While earlier studies still pointed to delays caused by AI tools, we are now also seeing clear acceleration. Tools such as GitHub Copilot, Claude Code and Codex have already become indispensable for many developers.

But these gains do not come without questions. Quality, maintainability, scalability and security are all areas where AI still struggles. At an organisational level, the picture is also uneven: many AI initiatives have yet to deliver measurable impact, while costs continue to rise.

AI has also become firmly established in marketing and design. Major brands are actively experimenting with AI generated campaigns. At the same time, opposing voices are growing louder. Discussions around copyright, authenticity and intentionality are more relevant than ever. Especially in more artistic domains, resistance remains strong.

Why product development is different

You might expect product development to follow a similar path. After all, we operate at the intersection of technology, design and business. Yet there is one important difference.

AI is only as good as the data it is trained on. And in product development, that is exactly where the crux lies.

While software code and visual material are available in abundance, the same is not true for physical products. Companies protect their designs carefully. CAD data is rarely publicly available. And even if you had access to that data, the question remains whether it would actually be usable.

Products are complex. They combine multiple functions, integrated solutions and countless trade offs between cost, manufacturability, user experience, reliability and aesthetics. True inventiveness is particularly difficult to train, because by definition there is no dataset of solutions that have not yet been conceived.

There are now impressive tools that can generate 3D models. But the gap between a visually convincing model and a manufacturable, reliable assembly is still significant.

Within CAD software, we are also seeing AI driven functionality emerge, such as generative design. These tools can be valuable, but for now they are mainly focused on creating individual parts. Not on designing complete, integrated products that need to function in practice.

Where AI does work

Does that mean AI has no role to play in product development? Absolutely not. I now use it myself almost every day.

But mainly for the low hanging fruit: translations, text editing and summarising. And perhaps most valuable of all: search. With chatbots, you can search much more precisely than with traditional search engines. They can help, for example, with finding suppliers, researching physical principles or data analysis methods, or mapping competing products.

The gain here lies in speed, without necessarily compromising on quality. Provided you remain critical, keep asking questions and verify your sources.

We also see applications in the field of design. Not by letting AI do the work, but by using it to support the preparatory phase. For mood boards, reference material, initial directions for materialisation, colour and finishing, and visualisations. As a result, the role of the designer partly shifts from creator to director.

Perhaps the most interesting application lies in exploring product propositions. AI can analyse large volumes of reviews and recognise patterns. What are users satisfied with? Where do frustrations arise? Which expectations keep coming back? It can help formulate USPs in relation to the existing market, structure ideas and visualise concepts at a level suitable for getting potential customers or stakeholders excited.

Over the past year, three clients have come to us who had refined and documented their product idea using AI tools. Not as an end point, but as a stepping stone. It gave them enough confidence to take the next step.

But that is exactly where the risk lies: AI can create the illusion that you are further along than you actually are.

The indispensable human factor

Because between a good idea and a successful product lies a world of work.

Translating a concept into a functional, manufacturable, reliable and attractive product requires more than data and algorithms. It requires trade offs. Context. Conversations with stakeholders. Solving conflicting requirements. And experience with materials, production processes, tolerances, use situations and real world behaviour.

Ultimately, it requires testing in the real world. Does the product actually work? Is it user friendly? Does it feel right? Does it hold up? These are not questions you answer with a prompt, but with physical prototypes, user feedback and iteration.

The human hand remains essential in that process.

Conclusion

AI is changing our field, but for now less radically than in software or graphic design. It is not taking over our work, but it is creating shifts. Just as CAD software, Photoshop and 3D printers have done before.

We are seeing a partial shift from creating to curating, and from making to assessing. For now, the real value lies mainly in supporting tasks: gaining insights faster, exploring possibilities, conducting source research, sharpening propositions and improving communication.

But the core of product development remains human work for now: creating things that do not just sound logical or look good, but function reliably in the real world.