How Machine Learning Will Make Businesses More Creative

Kai Goerlich

When I was a child, I used to love to lie on my back in the cool grass, the summer wind dancing across my face, and look up at the clouds as they drifted by. They changed form constantly as they moved, creating all sorts of creatures – everything from dragons, castles, and snakes to funny or terrifying faces.

If you think that this kind of image recognition is just child’s play, think again. Our ability to recognize patterns in the world around us is one of the most astonishing stunts that our eyes and brains perform. Whatever the material or form, whether it be clouds, rocks, or the surface of the moon, our brains instantaneously and easily “see” meaningful images within them, such as faces, animals, and structures.

However, there is a catch. While we are very good at recognizing patterns and images that nature thinks might be of relevance – such as the form of a potential predator in the woods – our processing power is limited. We’re not good at recognizing specific relationships among many different images. We quickly become overwhelmed, distracted, or simply bored.

Our brains on autopilot

The same goes for our short-term memory. We can hold only nine or fewer items in our working memory at one time. Even this number requires a lot of mental effort, as anyone who likes to play memory games knows.

We can blame our limited capacity on our prehistoric parents. Their brains became hardwired to think quickly rather than deeply to conserve precious energy and to avoid becoming dinner for hungry lions.

As a result, we are poorly adapted to a crowded, complex modern world. Even when we believe that we are making careful, rational decisions, the brain takes shortcuts that lead us to unconscious oversimplifications, biases, and an over-reliance on past experience in our decisions.

The ever-expanding heap of information available as an input to our decision making is not helping matters. It is overwhelming our already limited capacity. Complexity is robbing us of our ability to lie back in the grass and imagine.

With a little help from my algorithmic friend

Fortunately, machine learning is becoming exceptionally good at making sense of vast quantities of information without taking shortcuts. Machines already outperform humans in finding objects in digital images, especially at a fine-grained level, such as identifying not just a bird but the exact species. And machines can sift through thousands of these images without becoming tired, bored, or distracted.

This ability makes machine learning an ally in exponentially expanding our individual creative abilities in decision making. Consider fashion design, for example. Analyzing thousands of images to find the latest trends and hints for new ideas is a tedious task, but not to an algorithm. It can identify not just the latest trends in colors, styles, and accessories, but also convey the context behind all those elements, such as where they originated.

This may sound like it would reduce creativity, but ask designers, who flock to the places where other designers are, such as Paris or New York: Creativity builds on the ideas of others.

Inspiration also requires frequent feeding. Machine learning’s analytical capacity can provide all the food that our brains could ever want – in digestible form – helping to spark new ideas.

Whether it’s deciding on the next spring fashion lineup or making a complex strategic business decision, machine learning will free us to maximize our most precious human abilities: imagination and creativity.

This blog is the fourth in a six-part series on machine learning.

Learn more about the growing influence of artificial intelligence that reads and responds to our emotions; see Empathy: The Killer App for Artificial Intelligence.


Kai Goerlich

About Kai Goerlich

Kai Goerlich is the Chief Futurist at SAP Innovation Center network. His specialties include competitive intelligence, market intelligence, corporate foresight, trends, futuring, and ideation. Share your thoughts with Kai on Twitter @KaiGoe.