Everyone is talking about AI. But what does it really take to turn ambition into tangible business value?
In an episode of theBeyond Tomorrow podcast, recorded live during the Öresund 2025 Industrial Trade Fairs, Josef Forsman from Plantvision and Melina Katkic, CEO and founder of NordAxon, engage in a conversation that in many ways gets to the heart of one of the most crucial issues of our time. Not what AI is, but how we actually use it.
AI isn't new, but something has changed
The conversation begins with a bit of history. It’s easy to think that AI is a new phenomenon, but as Melina Katkic points out, the field has been around for a long time.
“AI is, after all, an old technology; the field was founded as early as 1956,” she says.
What has changed, however, are the conditions. For a long time, development was marked by so-called “AI winters”—periods when the technology failed to gain a foothold. The reason was simple: there was a lack of data, computing power, and the right technical platforms.
It is only in the last ten to fifteen years that anything has really happened. Breakthroughs in deep learning, access to large datasets, and the development of new architectures have created entirely new possibilities. When Josef Forsman asks what actually marked the turning point, Melina Katkic points to this very shift:
“That was when we were able to recognize cats and dogs in a picture; it was a very important milestone in the positive progress we’ve made today.”
When AI is reduced to ChatGPT, we lose perspective
An interesting point raised in the discussion is how AI today is often equated with generative tools.
“Now, you could say that many people equate AI with ChatGPT—that ChatGPT is AI. But AI is so much more than that,” notes Josef Forsman.
That’s an important point. Because when AI is reduced to a single technology or tool, we risk losing sight of the bigger picture. AI isn’t about a single application; it’s about a broad spectrum of methods and applications, ranging from image analysis to language technology and advanced time-series analysis. It’s only when we understand that breadth that we can begin to see where the real value lies.
Value doesn't start with technology—it starts with the business
One of the most key insights from the conversation concerns where to actually start. Because despite all the technological advances, the question remains fundamentally the same: What do we want to achieve? Melina Katkic describes it simply but aptly:
“We usually talk about what value we want to achieve. Do we want to increase revenue, reduce costs, or improve quality?” It’s only when we start with these questions that AI becomes relevant. Otherwise, these initiatives risk becoming exactly what many organizations are already experiencing: technological investments without a clear impact.
That's why so many organizations get stuck
The conversation then turns to one of the most crucial differences between those who succeed and those who don't. And it has less to do with technology than with people.
“The companies that truly succeed are those that have had people behind them who understand AI—or at least have been willing to understand it,” says Melina Katkic. This is an insight that is often underestimated. AI doesn’t require everyone to become an expert. But it does require curiosity, understanding, and active leadership. Organizations where AI is delegated without clear direction or expertise will struggle to create real value. Instead, it’s easy to get stuck in uncertainty or in solutions that never really take off.
Another important aspect of the discussion concerns how AI is actually implemented. There is a perception that AI must always involve advanced models and complex solutions. But the reality is more nuanced. In some cases, simpler, more rule-based solutions go a long way. In others, finely tuned models are required, or it’s necessary to build something entirely unique from the ground up. The key is not to choose the most advanced path, but the most relevant one.
As Melina Katkic explains, it’s about understanding both the need and the data, and then choosing the right approach.
The myth that data must leave the organization
A recurring concern in the industry relates to data security. Many feel that AI requires sensitive information to be sent to external cloud services. But here, too, the discussion reveals a more nuanced picture.
“You don’t have to send the data up if you don’t want to… you can use models locally on your own server,” explains Melina Kadic. This leads to an important insight: AI is not a threat to control, if implemented correctly. On the contrary, it can be combined with high security and full control over the organization’s data.
AI as a Support, Not a Replacement
Toward the end of the conversation, a more philosophical dimension is raised. What, exactly, is the purpose of AI? Melina Katkic is clear on this point:
“AI should be used to help people, humanity, the world, and our planet—not the other way around.” This is an important reminder in an era of rapid technological advancement.
Because, ultimately, it’s not about replacing people, but about empowering them. It’s about creating better decision-making support, more efficient processes, and more room for innovation.
From Hype to Direction
What becomes clear throughout the conversation is that AI in and of itself is not the answer. It is a tool. And as with all tools, its value depends on how it is used. Organizations that succeed are not necessarily those that invest the most, but those that start at the right end. Those that understand their current situation, define their goals, and dare to take steps forward with both curiosity and a clear direction. Only then is it possible to move from hype to real impact.
Would you like to hear the full discussion and gain more concrete insights from the conversation between Josef Forsman and Melina Katkic? Listen to the entire episode ofBeyond Tomorrow, recorded live from the Öresund 2025 Industrial Trade Fairs.