Can an A.I. system be a designer? (Part 1)
Examining the roles of the designer, creativity and technology in design
A colleague and I were having a discussion over the controversial topic of artificial intelligence (A.I.) systems in design. We were questioning the future of A.I. and reached the topic of creativity, that is, artificial intelligence that can create or design things that provide value for those that use them. We both agreed that neither of us wanted to see creative A.I. due to the risk of technological unemployment. The topic had me wandering in existential dread…
What does it mean to be creative? Is it just following rules like a coded program? To what extent is design creative? Are there binary rules for creativity? Does creativity come from memory or patterns? Ontologies? Taxonomies? Semantic queries?
If an A.I. can already link data to build meaning, then could it use that meaning to design creative solutions to problems?
What does Creativity mean?
Well I could just look up the definition in the English dictionary, but creativity can mean different things across people or cultures. Research centered on Hong Kong found that Westerners view creativity more in terms of the individual attributes of a creative person, such as their aesthetic taste, while Chinese people view creativity more in terms of the social influence of creative people e.g. what they can contribute to society. (Niu, 2006) In Scandinavian countries, creativity is seen as an individual attitude which helps in coping with life’s challenges (Smith and Carlsson, 2006), whereas in Germany, creativity is seen more as a process that can be applied to help solve problems. (Preiser, 2006) Twenty-three out of twenty-four African languages don’t even have an explicit word meaning creativity, yet people in these countries practice creativity every day. (Mpofu, 2006)
So we can deduce that there really is no universal definition or use for creativity, unlike the word “dog”- which means the same thing across cultures. I suppose creativity is in the eye of the beholder.
So what does it mean to design?
As designers, we are often faced with a problem that needs to be solved, or a system that isn’t working. It doesn’t exactly require creativity- but more of a deep empathic understanding of the context of the problem or thing. We learn about the environment before attempting to create for it. Creativity comes after the fact. So where an A.I. could technically “create” something, could it “design” something without human guidance?
I was reading an article by David Vandergrift on Medium who asked a similar question to the title of this article. He mentioned something interesting- a field called “computational creativity” (wiki), a multidisciplinary endeavour that is located at the intersection of the fields of artificial intelligence, cognitive psychology, philosophy, and the arts. They use examples of creativity like paintings and poetry created by A.I. systems as successes.
What’s notable about this technology is that creative pieces generated by A.I. are sometimes indistinguishable from human pieces- even by professional critics. Painters who created their paintings using AI began to have their work rated higher than pieces created by just humans. (source) This made me think back to the “chess cyborgs”, Steven Cramton and Zackary Stephen. In 2005 the duo left chess programs like Hydra, and grandmasters alike, in in their wake by using computers during a chess tournament. (source) This is a common example of an A.I./human (cyborg) combination being more efficient than just human or just A.I. alone.
Can an A.I. create an empathic bond?
I want to close this article by responding to the same philosophical argument that Vandergrift mentioned in his article: “an algorithm can only do what it’s programmed to do, so it can’t possibly do anything novel (i.e., be creative). Any product of the algorithm is solely a demonstration of the programmer’s creativity, not the machine’s.”
I partially agree with this argument. As a designer, I believe that both the designer and machine are equally accountable for the product. For the designer, it comes down to providing valid inputs to the A.I. via design research. For the A.I., it boils down to making the most out of the inputs and producing the best output.
Where a designer can build empathy, an A.I. can help create more questions, generate alternative solutions, or assist in building empathic understanding by providing feedback about user data. This allows designers to improve the frame of their ideation settings- and ideation, in it’s many forms, is an area today where humans have a comparative advantage over machines. (Brynjolfsonn & McAfee, 2016) Designers have used technology for years to accomplish their goals more efficiently, so why not artificial intelligence? So I ask- how can the combination of A.I. and human improve things for the people that use them?
Fabricio Teixeria provides a number of answers to this question in his article How AI has started to impact our work as designers. Facebook Designer Erica Virtue also discusses how she uses AI in her article Designing with AI.
My recommendation for a designer is to keep their tools sharp. Always consider how technology can help you be a better designer. As with chess, a partnership between an A.I. (e.g. Dr. Watson) and a human doctor will be far more creative and robust than either of them working alone. As futurist Kevin Kelly put it, “You’ll be paid in the future based on how well you work with robots.” (Brynjolfsonn & McAfee, 2016)
Next: Can an A.I. system be a designer? (Part 2)
Edit:
This Reddit thread has a nice collection of perspectives on this topic: https://www.reddit.com/r/Design/comments/7yn3my/can_ai_design_medium/?st=jdvtrpec&sh=3aebbced
Sources:
Brynjolfsson, E., & McAfee, A. (2016). The second machine age: work, progress, and prosperity in a time of brilliant technologies. New York, NY: W.W. Norton & Company, Inc.
Mpofu E et al (2006) ‘African Perspectives on Creativity’ in Kaufman JC and Sternberg RJ (eds) The International Handbook of Creativity p 465. Cambridge University Press ISBN 0–521–54731–8
Niu W (2006) ‘Development of Creativity Research in Chinese Societies’ in Kaufman JC and Sternberg RJ (eds) The International Handbook of Creativity pp 386–387. Cambridge University Press ISBN 0–521–54731–8
Preiser S (2006) ‘Creativity Research in German-Speaking Countries’ in Kaufman JC and Sternberg RJ (eds) The International Handbook of Creativity p 175. Cambridge University Press ISBN 0–521–54731–8
Smith GJW and Carlsson I (2006) ‘Creativity under the Northern Lights’ in Kaufman JC and Sternberg RJ (eds) The International Handbook of Creativity p 202. Cambridge University Press ISBN 0–521–54731–8
Vandegrift, D. (2016, June 09). Can Artificial Intelligence Be Creative? — David Vandegrift — Medium. Retrieved February 19, 2018, from https://medium.com/@DavidVandegrift/can-artificial-intelligence-be-creative-40e7eac56e71
