Why Do Designers and AI Need Each Other?

If AI is the new electricity, those who keep on designing improved oil lamps might quickly find themselves on the tail. That doesn’t mean we are to abandon principles of human-centred design and start every project from AI, but proficiency with the new toolset at hand and an ability to tackle problems by designing innovative data products will keep gaining in importance. Up to the point when looking the other way will become a losing strategy; Just as it is now, to overlook something like digitalization or mobile.

Designers and Ai - 1.jpg

If AI is the new electricity, those who keep on designing improved oil lamps might quickly find themselves on the tail.

Machine Learning Frame of Mind

Let’s call this a need to have a 'machine learning frame of mind', a solid understanding that allows designers and business leaders meet engineers half-way and conceptualize new solutions with how they stem from existing machine learning models with their tailored input, output and surrounding pipelines.

Innovating at Intersections

One way to gauge the possibilities is to stay in the loop with what is currently being done; Not only by the competition within the field but across the board in various industries.

Cross-industry innovation is often a catalyst to truly novel designs. Rather than pushing the boundaries in an established direction of improvement, cross-industry innovation intersects principles of existing solutions with conditions and constraints of the unique problem at hand and results in an innovative leap.

Designers and Ai - 2.jpg

This way of abstraction and reapplication is especially fitting for solutions within the data product realm where experimenting with semi-relevant datasets and model architectures is already one of the best practises for early-stage implementations. An impactful machine learning innovation on a product level does not necessarily stands on advanced or innovative approaches in the model architecture, there is a lot of potential stemming just from a clever application itself.

What Does This Mean in Practise?

We have to start with an understanding; Be able to mentally deconstruct what are we looking at to a sufficient level where we see the input, output and techniques used. From there we can deliberate how could the use case possibly translate into our given problem space, what are the foreseen challenges, performance and data requirements, potential pitfalls and strategy roadmap from the minimal viable product to up-scale.

Why AI Needs Designers?

Fortunately, the value and return of the investment of design from early project stages is, contrary to the past, widely recognised today. Iterations in design tend to be many times less expensive than pivoting further down the development pipeline.

Data products that emerge from design process methodology carry its benefits just as other solutions would. A good design helps us de-risk and devise innovations, products, features and new businesses that have better odds to catch on and successfully integrate into people’s lives.

Designers and Ai - 3.jpg

Why Now?

There are several myths surrounding AI and deep learning. Perhaps a baggage from the past that we like to carry as an excuse not to dip our toes into the somewhat intimidating subject. Too high costs, uninterpretability and a lack of owned data are some of the first few that come to mind. Machine learning today and five years ago are two different worlds; and while grounded in reality to an extent, those points are mute and with a number of approaches that help to overcome them. Be it prototyping, feasibility studies, return of investment evaluations or transfer learning for instance.

Whether you are a designer striving to deliver seamless experiences, a business leader looking for disruption opportunity and future proofing of your team's skillset, or an engineer arriving at the intersection from the other side: Start with small steps, and keep building up your 'machine learning frame of mind'.

We have the means to reinvent obsolete solutions and address problems that used to be out of reach for a long time.

Look around and see all the places where this is being done in this very moment.

And more importantly... the places where it's not.