If, like me, you were born in the 80s and attended a design school in the mid-2000s, you were probably taught some form of design thinking. Even if you’re younger or older, you have probably been exposed to design thinking at some point - perhaps even human centred design, depending on what flavour of design you studied.
For those not familiar with design thinking, it essentially boils down to a process of problem solving and critical thinking to establish a design that meets a client’s needs. This process is typically a series of reductive refinements, looking at the design brief in a wider scope and narrowing it down until the best possible solution is found. From the outset, design thinking is not a problem-solving methodology, but a solution based exploration.
It’s therefore not too far a stretch to argue that the more experienced an individual, the more efficient the design thinking process will be. The more years in industry an individual has, combined with a wider knowledge of existing work, enables that person to go through a process of design thinking with higher efficiency, as well as draw upon a wider pool of knowledge. This assumption isn’t unique to design thinking or designers, this is true of any industry.
Evolving design thinking
However, we are now living in the age of Artificial Intelligence (AI), generative tools, and smart products capable of creating feedback loops to inform future product design iterations. Combined, they are enabling engineers and designers alike to approach design problems in more innovative and efficient ways. Ultimately, this shortens the design and prototyping process and speeds up the time it takes for innovative products to hit the market.
Generative design creates thousands of possible solutions for a product – based on criteria and goals set by the designer –and produces geometries and forms that would normally take a human days, if not months, to develop manually. When combined with modern manufacturing processes, we’re also able to create products with structures that would never have been possible, even a few years ago. This process ensures the designer or engineer is picking the one design that meets the most important goals of their specific design. Embedding initial prototypes with sensors takes this one-step further – the feeding back of data essentially allows the product to co-design itself.
So, let’s bring this back to design thinking. Why couldn’t we, and why don’t we, develop AI tools to complement, if not entirely automate the first few steps of product design? What would that look like?
Enhancing the design process
Typically, any design process begins with a precedence study; looking to previous work, finding similarities to the current project, and analysing what worked or what needs more development. From there, you move onto finding common ground and building out concepts. These are developed in several different ways, but they all have one thing in common: key words. Any concept design is an expression of various grand visions that will help inform the final product. Finally, a few ‘low resolution’ prototypes are built – rough and ready versions that take less than 24 hours to develop – that communicate the idea. From there, a final brief can be created, and the project moves into product development.
Considering this process, it doesn’t seem unreasonable to incorporate AI and generative tools that can aid, if not completely take over, certain processes. Precedence study is essentially data gathering, something that AI is far more efficient at than humans. An AI program can be given inputs on what you’re looking for, for example sales figures with granular analytics, such as by purchases, region and demographic. These help to better inform the direction of the product. Imagine then cross referencing this against information on technology trends, or how the stock market is investing in certain technologies, and you have a really great foundation upon which to create a brief.
Next, conceptual level products can be generated using shape generator tools, such as generative design, which help give an overall aesthetic feel. However, it’s not unfeasible to imagine an AI system that could generate conceptual forms, based on cross referencing various existing products and using object recognition, to then extrapolate general geometries that can be used as a jumping off point for the design process.
A new collaborate partnership
So where does that leave design thinking, if AI were to takeover portions of the process? While there are often worries about technology making the role of designers and engineers obsolete, this is simply not the case. Design thinking isn’t going away, it’s evolving to take advantage of modern tools. In the same way a carpenter uses a power drill instead of a hand drill, the product designers and engineers of tomorrow will slowly abandon manual processes.
There are numerous examples of where these tools are already being used today, from house-hold items such as chairs, to the industrial design of car chassis and aeroplane partitions. AI and generative design tools’ capacity to drive innovation and efficiencies will see their adoption catapult across industries. They will become the new collaborators in the design process, enhancing the roles of the designer and engineer and creating products like we’ve never seen before.
By Paul Sohi, Product Designer, Autodesk