Business & AI /

Getting unstuck: How to adopt a creative mindset around AI challenges 

September 30/10 min read

The innovation that artificial intelligence enables is both a source of inspiration to leaders and a source of extreme angst. A lot of organizational leaders get mired down in the details and can’t quite envision how to bridge the gap to the future of possibility that AI brings. This “analysis paralysis” is an unfortunate state, because the time for adopting AI is now. Stall, and you risk getting left behind.

Bringing AI into your organization doesn’t have to be as life-altering of a concept as you might think. I spoke with Lars Ericsson, Head of Technology at Doberman, about what it means to adopt a creative mindset within the realm of AI, and how business leaders can convince themselves and other stakeholders to start moving on this very important transformation. Here are a few of the takeaways from our conversation.

Getting unstuck

Ask yourself what problems you have, not what AI can do

With the advancement of deep learning, you can now use data you already have in brand new ways — even data you did not know you had. But a lot of leaders get stuck trying to wrap their minds around how to use AI technology. It’s that focus on technology that is the mistake. A better approach is to ask yourself, “What problem do I have that needs fixing?”

A great example of how AI can solve problems is the story of the manufacturing floor, where a dedicated machine operator has developed a symbiotic relationship with his machine. He is so attuned to the sounds the machine makes that he can detect, with uncanny accuracy, when something is off and the machine needs a tweak or tune-up.

“If only we could scale that,” the plant manager muses. But he can. The sounds and vibrations machines emit are a type of data that can be analyzed by deep learning to find patterns over time. AI can get just as good as a longtime machine operator at predicting maintenance — better, even, and in a scalable way.

Ericsson says: “If you step into simply optimizing what you already have or already do, you’re not going to get very far. A lot of the times, what we try to do with customers is to take away that limited mindset. Instead of starting with how to optimize a process, we talk about the problem that exists that needs to be solved. If you start with that, you don’t get hung up on blockers that get in the way of the creative process.”

Use filters when you discuss AI

Applying AI to sound data might not occur to the business leader who takes for granted that “only people can do some things.” But if you put words to your challenge, you might be surprised what AI design can accomplish. Ericsson calls this “the superhuman question.” If you had a superhuman, what could they do for you?

“Reality is a barrier for creativity,” says Ericsson. If you’re trying to apply a technical “AI lens” to your business problems, you’re not going to get very far. 

The superhuman question gets you away from talking about AI in technical terms, particularly when discussing it with other stakeholders in your organization. The barriers to imagining AI can be broken down if the conceptual elements can be discussed in real terms.

This helps you be more conceptual about brainstorming problems upfront. For instance, these might be some problems you’re looking to solve:

  1. Improve revenue
  2. Reduce costs
  3. Expand into a new market

They’re all pretty high-level universal problems, but they provide a good starting point. To consider around what AI could do for your organization, it’s important to abstract your thinking, letting potential features go for now. From there, you can start to drill down to a tangible use case to start with.

Show, don’t tell: Prove that AI can work

Don’t spend too much time finding the right first application for AI, and by all means don’t start with your biggest, boldest, hairiest business problem. Instead, find a smaller, more containable use case so you can prove quickly that AI works. “That’s a great way of unlocking the filter,” says Ericsson. 

Another way to look at this: Instead of tackling your biggest problem right off the bat, focus on just one aspect of that problem as a first step. For instance, for insurance companies, AI holds the promise of telling them exactly how much to pay out in each claim. But this is a huge paradigm shift. As a starting point, insurance companies could use AI simply to classify the damage done via images. (Peltarion has a car-damage classification tutorial that does just this.) AI can determine whether the damage shown in an image is a headlight, a door, or a window, for example. 

Don’t start with your biggest, boldest, hairiest business problem

Once this use case is put in motion and deemed successful, the next might be to raise the detail on the classification: Is it the driver’s side door or the passenger’s? Is it a headlight or taillight? 

“It’s okay to tackle big problems,” says Ericsson. “But start with a simple solution and iterate. Take it step by step.”

Expand your mindset

A lot of people think that implementing AI is about mastering technology, but it’s really much more than that. It’s about changing the way you think. When social media arose, people scoffed at its usefulness. Now look at us. The same can be said for “the mobile push,” which didn’t just allow customers to access websites on phones, but created a whole new business model for apps and changed the way nearly every company does business.

AI is the same. It isn’t just going to help us solve persistent business problems in a new way. It’s going to change the way we interact with our customers and invoke massive organizational change within organizations. And that can be scary. Which brings us to the next point.

Create an outlet for the anxiety

The third best practice for adopting a creative AI mindset is to deal head-on with the inevitable anxiety that arises around AI discussions. Innovation often creates anxiety because people worry about how change will affect their roles and projects. Rather than being encouraged by possibility, they panic. Getting people ready for change means managing that anxiety and giving them an outlet to vent.

When you’re in an anxious mindset, you’re not at your most creative. You need to turn your anxiety into creative excitement. Inspiration can work well for this. But sometimes you need to lower expectations a little bit. A lot of the AI anxiety comes from a misbelief that there’s a lot of pressure. You need an outlet for your anxiety, and it can be as easy as just letting people articulate their fears around AI.

Stall, and you risk getting left behind

The speed of development of new tools and technologies is so much higher than the adoption speed, which creates a bigger and bigger gap between what’s possible and what people are actually doing with technology. The mission of both Doberman and Peltarion is to close that gap. Make it easier for people to understand what there is to do, and to implement it. 

Keep your eyes on the AI prize

You probably get a lot done with your smartphone without understanding exactly how it works under the hood. And you don’t need to be an expert technologist to use AI, either.

In fact, Peltarion’s operational AI platform was designed to help all types of leaders and organizations get deep learning off the ground — not just data scientists. So the first step to getting started with AI is, frankly, simply to get started with AI.

Getting started playing around with the materials is a priceless part of the creative process. Experimenting, hacking and prototyping are all valid ways to experience AI technology and get into a comfort zone with using it. We’ll expand on this in the next part of this article series. In the meantime, get started with AI now and deploy an operational AI model with this tutorial

  • Reynaldo Boulogne

    Reynaldo Boulogne

    Head of Partnerships

    Reynaldo is Head of Partnerships at Peltarion, Co-founder of the Nordic AI Alliance and Vice-chairman of the Stockholm based AI forum, Stockholm AI. With over 15 years of experience, Reynaldo has worked within the intersection of business and technology across multiple sectors, most recently at Klarna and Spotify. He is passionate about innovation, leadership and building things from scratch.

02/ More on Business & AI