Can taste be taught?
And can an artificially intelligent system be the one that teaches it? It’s a question we ask ourselves every day as we continue to build a online logo maker and graphic design platform from scratch.
But the future of AI is exactly in this kind of subjectivity.
AI is already moving towards more qualitative decision-making, as it continues to make strides in the worlds of art and design. Algorithm-generated art even made headlines when a work by AI “painted” art sold for over $ 400,000 at a Christie’s auction. While this painting is just one example, it raises the question of whether AI deserves a place in the realm of artists and designers.
Data scientists and engineers working on this question may well have an answer. If AI is to work alongside designers, it must understand the principles of design.
This solution itself raises some problems. To know how exactly will you teach these design principles to AI? Teaching an AI-powered design platform to make better design decisions shows us how these principles could one day form the basis of an artificially intelligent designer with their own artificially intelligent sense of taste.
Before we can discuss the process of teaching design principles to an AI, we need to discuss exactly what these principles are.
Contrary to popular belief, designers don’t run entirely on instinct. Behind all the split-second decisions about what looks good and what doesn’t, most designers can explain these feelings in a logical way, using principles of design.
Not to be confused with design elements, like color, shape, texture, and line, design principles are about the arrangement of these elements. When designers assess whether something “works”, these principles provide the framework for judging the quality of the visual design.
They understand balance eemphasis, movement, pattern, repetition, rhythm, variety and unity. In the same way that designers use these principles to guide their decisions about what looks good, so could AI learn to do the same.
Obviously, getting the AI to understand design principles isn’t as easy as saying “Look for contrast, emphasis and unity in this work”. You must teach him what these concepts mean. But by breaking the design principles down into a series of smaller decisions, AI could learn them from scratch.
How an AI algorithm makes design decisions
While we’ve mostly looked at AI and design on a theoretical level so far, this idea is far from pure theory. Today, AI is already learning to make the kind of complex design decisions that form the basis of design principles.
Let’s take a look at some design algorithms in action.
In all of these examples, the AI in question taps into data from hundreds of popular logos across different industries and finds commonalities in what makes those logos beautiful. By isolating different trends in logo design, AI can learn to apply these rules to new logos as they are created.
In short, we’re watching AI create its own set of design rules to work on.
In this first design, we can see the algorithm choosing to couple the style of the symbol with the style of the font. Among popular logos, this kind of continuity between weight, rounding, and color for symbols and fonts is a popular way to make a logo more cohesive – something the AI has learned to apply.
|Advice: Read this article to learn more about common symbols in design|
This type of coordination is also an aspect of the design principle of repetition. The repetition of certain elements makes a room appear more like a unified whole, as its parts appear to visually “go together”. As the AI examines more designs, it will be able to understand and apply more types of repetition, including complex designs like patterns.
For this logo design, the algorithm picked something that is both practical and attractive, and that is the trend of logos with long names stacking up words. But beyond learning that this particular business name needs to be stacked according to design tastes, it also needs to understand how to properly organize words and space letters.
Together, these choices confirm the design principle of accentbecause the stacking of the company name contributes to the visual hierarchy of the logo. By focusing on the name of the company rather than the tagline, AI learns how and what to showcase in a design.
When clients ask designers to make things ‘pop’, they are usually referring to the design principle of contrast.
Here, the algorithm begins to generate contrast in designs by selecting background and font colors that meet a minimum contrast value. By setting limits on what it sees in the popular logo design, the AI can decide which color combination is most likely to grab the viewer’s attention.
While the AI here has learned to make a diverse set of design decisions, the rules it created based on popular logo designs only touch small areas of much larger design principles. But as AI begins to learn from the sum of all these self-taught rules, design principles could begin to guide the work of AI-powered designers so they can move beyond to suggest To to create.
What AI-generated symbols tell us about design
Making a few design decisions is one thing. Creating a design from scratch is quite another thing.
To create cohesive designs, these decisions (and the principles to which they relate) must work together, some taking precedence and others being behind. This is where things get interesting.
The AI-generated designs will show us exactly what all of these decisions look like when we work and compete with each other.
Take a look at these four hearts:
What do they all have in common?
If you guessed that they were all created by artificial intelligence, you are right.
While it might sound simple, creating this kind of variation isn’t easy. As the design sensitivity of AI continues to improve, symbols like these will only get better. But that doesn’t mean they aren’t already doing well.
Indeed, people who use a logo generator choose AI-generated symbols like these about 50% of the time when given the chance, which makes them just as popular as human-made symbols. Not too bad for a system that is still learning design principles.
What can we expect for the future of AI and design?
While we’re still a long way from artificial intelligence which can create complex, human-made designs, teaching AI principles of design is how we’re going to get there. As machine learning algorithms incorporate more and more information about what makes a good design, eventually they will begin to understand the principles of design.
It is through these broader principles that AI and designers will eventually become indistinguishable. Because whether these principles were learned in a classroom or through code, the same rules still apply.
|Related: See how AI and embedded intelligence work together to advance technology by integrating software with small devices to improve everyday life.|
Want to immerse yourself in everything related to graphics? Browse our content hub with over 50 unmatched graphic design resources.