The real problem with the interface is that it is an interface. Interfaces get in the way. I don’t want to focus my energies on an interface. I want to focus on the job… I don’t want to think of myself as using a computer; I want to think of myself as doing my job.  

This is how design guru Donald Norman succinctly described the future of user interface (UI) and user experience (UX) design way back in 1990 in his book, The Invisible Computer.  Surprising to say, almost three decades later, we are making this a reality.

With Artificial Intelligence, chatbots, voice-enabled devices, natural language processing, and machine learning have taken the center stage. We are in an era where designers must go beyond screen-based thinking, demanding immense time, money and talent to make. The shift to a “No UI” and data-driven design path has begun. UI/UX designers must craft smarter and more useful systems that go beyond the inherently unnatural interfaces of the past that is painstaking to use and not to mention diminishes value over time.

Invisible User Interface

Artificial intelligence such as chatbots and voice controls like Siri and Alexa are moving user interface designers beyond the screens, and into invisible UI design. Also known as Zero UI, natural gestures, voice, facial features, and biometrics are gradually taking over the mainstream of device communications.

In fact, Voice User Interfaces (VUI) have already reached critical mass. This means VUI will continue to grow, eventually turning into a household name. Just take the case of Amazon Echo and Google Home—both techs have increased its adoption rate to 24 percent in the last quarter of 2018.

In designing for invisible user interfaces such as VUI, it is important for designers to use familiar voice patterns to launch, navigate and use an application. Designing for invisible UI means using universal conversation patterns in executing commands that are typically short and sweet. This will help lessen the cognitive load and prevent user confusion that may lead to them abandoning the application.

Thin User Experience

A User Experience is thin when the customer only needs to perform a few steps to get value from a product. For example, purchasing online has a thinner user experience compared to going to a physical store. Instead of driving to a shop, browsing an aisle, and waiting for a long line to pay; customers can make a few taps on their mobile phones and settle their bills in a matter of seconds.

A thin user experience can be more efficient through anticipatory design. Artificial intelligence can help designers anticipate patterns of buying behavior, product preferences, and even feature consumption. Machine algorithms can help predict customer requirements from data patterns based on previous customer interactions. In turn, UI/UX designers can use these AI-defined requirements to consistently improve their front-end and deliver a thinner and faster user experience.

Design for Interaction

Applications built under an artificial intelligence backbone improves its algorithms the more someone uses it. Machine learning models rely heavily on human input and feedback, to regularly improve its patterns and formulas to provide better insight, prediction, and recommendations.

Just take the case of Tay—Microsoft’s notorious chatbot gone rogue after Twitter trolls bombarded it with racist and sexist Tweets. The algorithm behind Tay took user inputs to independently update its patterns and computations, which unfortunately led to a mean AI.

User experience designers must, therefore, design for an interaction between the machine learning model and its users to provide not only constructive but also direct feedback. A good example is placing a simple thumbs-up or thumbs-down icon at the bottom of each output to give a quick and simple reaction of how the algorithm works. These positive or negative responses can be generated regularly and can be used as input by Data Scientists to further improve their algorithm.

Continuous A/B Testing

Designers often use split testing to know which version of their designs works best with the customer. Typically, this means producing two versions of a website or application and have users from various demographics use it. The challenge though with this approach is that UI/UX designers only get insights from the needs of the highest-ranking user types, easily neglecting the needs of others.

With artificial intelligence, designers can replace A/B testing with machine learning algorithms. Take the case of Netflix. The video-streaming turned movie-producer mogul, anchored data-driven decision making in its backbone. Through big data gathered in real-time from millions of user interactions, Netflix not only can continuously update its design relative to user preference and feedback but also effectively recommend content not just to certain customer segments but to the specific individual subscriptions.

AI-powered User Journey

Remember how painstaking it is to create a user journey map? Countless hours, discussions and arguments just to determine what will the customer do next. Artificial intelligence is not just providing designers arsenals for data-driven decision making—it is even updating UI/UX designer tools to make a truly intelligent front-end.

For example, tools such as  ReFUEL4  can use predictive analytics to grasp the online journey of customers and plan them into sections based on their choices and preferences. Designers can start using AI-backed tools such as ReFUEL4 to cut-down hours spent on understanding user requirements and focus more on their core creative expertise.

Key Takeaway

Chatbots, voice-enabled devices, natural language processing, and machine learning have changed the game in UI/UX design. It is time for designers to go beyond “screen-based thinking” to a “No UI” and data-driven design. Crafting smarter and more useful systems that rise above the inherently unnatural interfaces of old graphical user interface, requires creative minds that not just make front-end solutions. UI/UX designers must shift their focus from solving problems, to crafting a system that adapts not for computers, but for people.