Episode description

The O’Reilly Design Podcast: AI, understanding algorithms, and design diversity.

In this week’s Design Podcast, I sit down with Giles Colborne, designer, author, and managing director of cxpartners. We talk about how AI is reinventing design and the roles of designers; the balance of creating something that is different but familiar; and how, at its most basic level, AI is shortcutting user input.

Here are a few highlights from our conversation:

The trouble is, there's a dirty little secret there, which is mobile; it's the platform that people want to use. If you give them a choice, they'll reach for their smartphone every single time. It won't work very well. You look at eCommerce sites, the average conversion rate on an eCommerce site is something like 4-4.5% average globally. On a mobile site, it's less than 2%. Although the conversion rates are rising, they're rising at the same pace. Usability of desktop is getting better, but at the same pace as mobile.

That's terrible. That means a lot of businesses are seeing their traffic shift to a channel that actually doesn't work as well, but people would like it to work well. At the same time, these devices have become incredibly powerful. At the same time, organizations are suddenly finding themselves flooded with data about user behavior. Really interesting data. … The interest in AI at its simplest, the crudest application of AI, is simply to shortcut user input. That's a very simple application, but it's incredible powerful. It has a transformative effect.

What's interesting is, in talking to a lot of people who are incorporating this into their design practices, themes are emerging. Amber Cartwright at Airbnb said a very simple, profound thing, which is if you want to design a thing, you have to understand it. If you're going to design with algorithms, you have to understand them. Like all of us, she's running a design team. She can't just send everyone off for three years, get them a degree in higher mathematics, and we'll see you when you've done that.

She did the sensible thing, which is she said, "We're running mixed teams. I'm going to put the design teams and the data scientists next to each other, and I'm going to get them to draw out what it is they're trying to do." That's a very simple thing, but immediately, she started to see people sketching out, "This is what this looks like. This is how this works." By drawing it, the abstract becomes concrete. By drawing it, there's a common language. By drawing it, you get that spark.

On one hand, you have this globalization of design. Through the network, design patterns spread incredibly quickly. That leads to a great sameness in design. You see that when Apple or Google releases something; that pan spreads incredibly quickly and ends up everywhere. That's good, because it means that there's a familiar common language.

The challenge of design is, how do you have that familiarity and common language and respect preserve the diversity of design? People are going to, at some point, react against that, and they're going to say, "I want this to be different, but I want it to be familiar, too." I think diversity in the design output, diversity in the design community, also, is incredibly important. That's the input, if you like, to the design output.