Design + Sketch App — Medium | Cole Quartuccio

Starting our Journey

What you’re about to read is a vision of future AI through the lens of a tentatively optimistic UX Designer. Just to be clear, I have no mathematical or data science street cred of any kind. I have a YouTube level education in this stuff, but it’s so fascinating I can’t keep from chiming in. This article is part one from “UXD meets AI”, a three part series. Peep the other articles here(Coming Soon): “UXD meets AI: Part 2 — Surveying the Landscape”, “UXD meets AI: Part 3— Getting Cosmic”.

I’m inviting you: fellow designers, product managers, developers, data scientists and technologists on a journey to understand the impact of Artificial Intelligence(AI) on the creative process. In the “UXD meets AI” series we will hypothesize about our partnership with AI as a design tool, uncover existing creative algorithms and ponder cosmic questions about AI that will undoubtedly become part of our daily orbit as we approach the singularity.

First things first, let’s look into the future. Part 1 of this series is my personal exploration of a concept for an artificially intelligent design tool. It’s also a call-to-action for you. If you make it to the end of this article — godspeed — there’s no doubt that you’ll have some thoughts or feedback of your own. I want to hear all of your brilliant ideas.

Meet Cyclo™ — Your AI UX Design Tool of the Future

Imagine you’ve just shipped the first iteration of a new feature. Woot woot! Once you and your team are done poppin’ bottles, making it rain and basking in the affectionate glow of 5 star reviews, you decide to take another look at your analytics to see how the feature is performing. After reviewing usage data and qualitative feedback from customers, your team decides to rethink how users add people to a group. Here are the data trends you have to work with:

• Usage Data: On average users add about 35 people to a group. Wow, that’s way more than you thought.
• Qualitative Feedback: Users are saying they “just can’t deal” with how many steps are required for adding people to a group.

“Well isn’t that special!?” you think to yourself while plugging their feedback into Cyclo, your trusty design iteration software program.

👨‍💻 You: Inputting data and feedback

UI Task: Add people to a group

Usage Data: Avg. amount of people added = 35

Feedback: Reduce the amount of steps required to add a person to a group

🤖 Cyclo: Thinking… [robot voice]

“Bee, doop, beep, beep, computing, computing. Cy-cles Com-plete.”

👨‍💻 You: Receiving new iteration

“Damn Cyclo, that was fast!”

🤖 Cyclo: New iteration [robot voice]

“Use an index list view of users on the platform. Display users first and last name with a “+” icon to add them to the group.”

add-users-v1.cyclo

Oh look, Cyclo is asking you a question!

🤖 Cyclo: Asks Question

“Are we adding users from inside or outside of the platform?”

👨‍💻 You: Answer

“Why yes Cyclo — these users are inside and outside of the platform.”

🤖 Cyclo: Making some tweaks… [robot voice]

“Bloopity, bleep, bleep, bee, doop. Cy-cles Com-plete.”

add-users-v2.cyclo

While looking at the new iteration, you notice Cyclo has updated to a more search-focused ui. With this new approach users can more efficiently search the index, invite users outside of the platform via email and review their user selections in aggregate. Even the helper text has been updated to “Name or email”.

👨‍💻 You: In Jaw Dropping Amazement

“Nice touch Cyclo!”

A solid day’s work

Would you look at that!? Cyclo created an interface for adding large amounts of people to a group while enabling users to perform the action in fewer steps. It looks like you have a promising design you can start testing with users. Although design is a creative pursuit, at the end of the day we are responsible for solving problems, which are composed of modular objectives and constraints. Just like building something with Lego pieces a designer is organizing, stacking and breaking apart the artifact until they’ve reached an intuitive and impactful build for their users.

The real challenge will be teaching artificial intelligence to understand company objectives, design constraints and the nuances of human experience that guide our design decisions. In it’s early iterations Cyclo could be more rudimentary by relying on the designer to fill in the knowledge gaps. After riding shotgun with millions of ux peeps, Cyclo would be in a good position to develop situational awareness for common design tasks. To be fair, building Cyclo may only require machine learning. We don’t necessarily need this thing to pass the Turing Test, but you can think of it as more of an narrowly intelligent tool or ANI.

What does UXD meeting AI look like for you?

Let’s keep in mind, nothing about our interaction with Cyclo required a mouse, keyboard or any kind of physical input for tweaking designs. In other words, we could go straight Westworld on everybody’s ass and audibly engage with our design artifacts. Who’s to say that Cyclo is limited to the narrow use case above? If Cyclo can learn the entire product development process it could:

• Lay out the information architecture of your app
• Vet plausible design patterns
• Recommend alternative user flows
• Prototype your app based on the functionality of the UI patterns
• Synthesize qualitative and quantitative feedback from customers
• Suggest prototype tweaks based on usability patterns
• Automate A/B testing with awareness of conversion goals

Basically anything we can do, Cyclo could do better. How do you think Cyclo could help you in your UX Design workflow?

Tune in for the next article where we scour the landscape of creative artificial intelligence to vet the technical feasibility and product market fit of Cyclo — our trusty design partner.

UXD meets AI: Part 1 — Discovering a Workflow was originally published in Design + Sketch on Medium, where people are continuing the conversation by highlighting and responding to this story.