Augmented Innovation: Using AI to bring back Corporate Creativity

As we were starting the second decade of the 21^st century, it had already become a growing concerning, that nearly half of innovative work that made their way to patent offices or media were nothing notably beyond slightly modified copies of existing ideas.

Compared to 100 years ago, we are spending over twenty times on innovation, while getting much less out of human mind in sense of pure creativity. That is despite all of the new age “thinking outside the box” approaches in our pursuit of new ways to intrigue the mind and expanding our perspective about the problems we want to address.

To make the current status even more critical, it is estimated that over 95% of patents are not licensed nor ever make it through commercialization as a successful solution.

It seems that human brain’s capacity in combining existing ideas, or their key components, and to create new ones is flatlining on us.

Our minds are stretched so thin between the sources of daily noise, which are caused by numerous social media outlets and digital distractions, that even achieving a strong focus to allow for playing LEGO with existing ideas and building new ones with them is getting out of human reach!

AI joins the R&D department

Artificial Intelligence (AI) has already been expanding in every aspect of our lives, serving us in a daily basis, visibly and behind the scenes.

From a seemingly simple mobile app that connects us to a social media platform or shows us the weather forecast, or voice commanded digital assistants, like Alex or Siri, to the end of the entire supply chain that runs our industries and bring our daily needs to us, all is now very dependent on AI to serve us with an ever increasing efficiency, accuracy, and relevancy over time.

It was only a matter of time before our innovators would try using AI in skimming through existing ideas (or registered inventions) and try to come up with workable combinations that would be “creative enough” to be able apply for its own patent and be “close enough” to a solution for a real world problem.

One good example is the Analogy Miner developed at the Carnegie Mellon, which uses algorithms to find analogies based on “problem schemas” to identify the purpose of a patented product (or invention), and another real life phenomena. 

Instead of looking for completely fresh ideas, the focus will be in coming up with completely fresh connections between concepts, that would in turn respond to a real life need and be reasonable and feasible enough to be put into experimentation.

Analogies are very important in driving innovation and many important inventions and discoveries in history are driven by them. Perhaps the most famous one, was the apple that fell in front of Sir Isaac Newton when he had taken refuge to a rural estate away from a pandemic hit London, which triggered the connection between the falling object and the math behind the formulation of gravity.

Another good example would be the analogy to a bicycle that gave the Wright brothers the idea to create the steering mechanism of the first airplane in history.

The Augmented Innovation (or as some would call it, Augmented Creativity), is the process where we use AI to help us better envision ideas out of the existing massive amounts of data that we have at our disposal today.

Over the past recent years, we have seen quite fascinating examples of AI driven creativity in arts and literature as well.

In 2016, Microsoft, in collaboration with ING and University of Delft, used AI to study several work of the famous Dutch painter, Rembrandt (who died 4 centuries ago), and use the same approach, attention to details, paint brush strokes, color tones and even color pigment granularity, to create a new work matching exactly how the original artist would have done it.

The outcome, which was later titled as “The Next Rembrandt”, had such a great authentic implementation that a group of art experts, who did not know the origin of the work, agreed that this must be a missed original work that has been hidden for centuries.

AI uses Big Data (i.e., massive amounts of data) for this mining activity. To do that, we use the most common algorithmic approach in Machine Learning (ML) – which itself is a sub-group of AI – named as Generative Adversarial Network (GAN).

These models are capable of learning from humans by observing how it is done by them, and then practice internally to get better at doing so, and in quite a number of cases, they surpass human’s brain accuracy levels.

OpenAI, a San Francisco-based AI research lab, came up with GPT-3 in June 2020. This model, uses 175 billion parameters and have been trained by going through hundreds of billions of rows of training data, and is capable of writing poetry, stories, movie scripts, computer codes and even your resume, with such a high quality that is impossible to distinguish whether it is written by a human or a machine.

Gaming and Movie industries have been accelerating use of Augmented Innovation to help them with creating attractive stories and ideas to engage their customers in a very distracting world. IT has been especially tough to create interesting content that can absorb and hold viewers sitting behind their computers, while having multiple browser windows open, with their smart phone next to their keyboard.

Democratizing Augmented Innovation

As any enterprise which has successfully implemented their Digital Transformation would tell you, one key aspect of Digitalization is to continuously adapt the Value Proposition Model with the dynamics of the shifting market landscape.

That is why we now see efforts in the market in trying to turn their internal ML-enriched Innovation tooling into a platform-as-a-service working model where their subscribers can use that to look for valuable analogies or idea-combinations that can be mapped as a solution to their client’s problems.

We can expect that in near future, these platform will become available to the general public for their own home-based creativity practices.

Conclusion and Caveat

The considerable boost that Corporate Creativity has received from AI Augmentation, comes with a hidden price tag, as the main problem with human brain’s flatlining on true creativity persists and can even get worse by relying further on machines to do the innovative thinking for us.

It is true that market needs new and relevant products, services, and content, but as we move away from direct involvement of human ingenuity towards more machine produced ideas, we may start creating a drift from what makes us humans, towards delicate hidden biases that may exist in our algorithmic models.

This would be especially difficult to detect and rectify as we use Deep Learning as our core Machine Learning practice for innovation, which makes it nearly impossible to explain and track to the logical that is used in generating the ideas.

In incorporation of Augmented Innovation with Human Creativity, special, ongoing care and discipline must be practiced to maintain the contribution balance between the Organic Intelligence and the Artificial Intelligence, to keep both players in the game in a true form of Augmented Intelligence.