88% of designers surveyed believe that it will be at least 5 years or more until visual designers are replaced by AI. AI can already do a lot right now. —John Maeda, 2018 Design in Tech Report
Designers are frightened.
Some of them.
I think I understand why. I’ve read a book by Andrew McAfee and Erik Brynjolffson called Machine | Platform | Crowd—the seemingly inhuman punctuation of the vertical lines in the title (|||) seem to radiate a warning to the people in the audience! The volume’s fifth chapter, about technology and humanity, blithely informs us that “machines are getting quite good at coming up with powerful new ideas on their own.” The book talks about something called “generative design,” a kind of software that draws, does cost-benefit analyses, and crunches the necessary numbers all by itself—producing designs that satisfy exactly what the briefs request. You can imagine a captain, or even a sergeant-at-arms, of industry reading these words: “…unlike most, if not all, human designers, the software isn’t consciously or subconsciously biased toward existing methods, so it really does explore more freely” and rubbing her hands together.
To prove their point, the authors talk about a bizarre-looking heat exchanger produced by generative design, which looks like metal LEGOs melted and twisted by an angry, superhuman toddler. They try to assure readers that “Humans played no role in designing the part ... But they were essential for telling the generative-design software what kind of part to design.”
This is of limited reassurance.
M|P|C’s contention is that, in the future, humans won’t design as they had previously, but will instead be chiefly responsible for feeding the appropriate data into a super-smart design machine—and perhaps checking the output to make sure it makes sense to, you know, people. Which is to say: Designers will essentially become… programmers? Super-specialized data-entry folks? The fanciest QA professionals ever? Whatever the proper term, one wonders if such a shift will work for designers. How will it make designers feel? How fulfilling, how much fun, how interesting will such a job be? Design is a meaningful activity. It seems that AI is threatening that. (Nota bene: Had the authors talked about the improved value and earnings potential, the professional fear and trembling might have abated for some designers.) Since designers choose their profession because it’s intrinsically motivating, not because it necessarily pays the big bucks, this rejiggering of the discipline could result in a serious diminishment of autonomy, mastery, and purpose.
M|P|C’s contention is that, in the future, humans won’t design as they had previously, but will instead be chiefly responsible for feeding the appropriate data into a super-smart design machine—and perhaps checking the output to make sure it makes sense to, you know, people.
Of course, design isn’t just about heat exchangers. So much of it is about identifying and satisfying human needs and wants for a wide variety of products, services, and experiences. For instance, by talking to moms, and listening closely, and identifying their emotional focus on childhood development, Continuum was able to create Baby Stages of Development. The truth is, we still need people who understand what is humanly moving about a particular design. AI will whomp us when it comes to speed and pattern recognition. That’s indisputable. But it’s important to remember that we see things that computers don’t—at least not for the foreseeable future. Design requires someone who can truly account for human context. Seeing subtle social and emotional realities in products and services isn’t something you can necessarily get from a dashboard.
Does this mean that the fear that AI will rob design of its special-profession status is something to ignore? No, it’s real. But perhaps we need to help designers see it less as an apocalypse and more as a professional evolution. AI is a tool, a really powerful one, for designers to use; not an alternative to designers. Not now. Imagine, if you will, the fear and loathing some old school designers might have felt when, say, Photoshop swaggered onto the scene way back in 1988!
I’d suggest, for those looking to be talked off the professional ledge, reading the piece that Rodney Brooks, legendary roboticist, recently published in Technology Review about AI-prediction hysteria. In it, he notes something called Amara’s Law: “We tend to overestimate the effect of a technology in the short run and underestimate the effect in the long run.”
AI is a tool, a really powerful one, for designers to use; not an alternative to designers.
The idea of replacement by AI is the problem. It seems much more reasonable to say that in at least five years, designers will be regularly using—or, if you prefer—collaborating with AI. A terrific example of this can be found in my colleague Buck Sleeper’s essay on the benefits of what he calls “data friendship.” The integration of AI into our professional lives might redefine what makes for a good and valuable and employable designer. Is this a reason for panic? It becomes less so, if one chooses to learn what’s going on, rather than just lament the changes. Read M|P|C. Take a class on automated design. Find ways of getting your hands on these advanced technologies and learn to use them. Make for yourself an adaptive mindset.
Those who insist on maintaining a sentimental attachment to the ways things have always been done may well find themselves in professional trouble, as things evolve.
Interestingly, M|P|C can guide designers to a more open mindset. The book asks questions such as—
- Where would better human connections most help your performance and that of your organization?
- Looking at the existing tasks and processes in your job or organization, what do you see as the ideal division of work between humans and machines?
- What new products or services could be created by combining the emerging capabilities of machines with a human touch?
These questions can, if taken seriously, help designers and the people who employ them reframe the issue. We need to engage in such Q&A if we’re to remake design as a worthwhile activity in the future. Designers must learn how to make the argument for their own value and inclusion: to technologists who make the tools, and to the clients and bosses who may or may not hire them. This may sound daunting, but it’s surely going to be necessary. So let’s put our intelligence—our natural intelligence—to work.
