Can Artificial Intelligence (AI) become a creative friend?

Blog Post by Lisa Bowers & Elouise Huxor

In his article ‘Crafting an Artificial Intelligence I‘, Yatharth argues that there are parallels to be drawn between the craft process and the development of machine learning through engagement with the data as material. The word technology has its roots as a Greek word roughly translated as ‘a Discourse and Treaty on Art and Craft’, the distinction between the terms technology and crafts is something that emerged with the rise of capitalism and industrialisation, it is not inherent in the word itself. (  Figure 1

Moving towards technology or AI as a friend, rather than foe, of the creative practitioner does appear to depend on your perception and need. The consideration of technology now is not so much about a cold mechanical monster which will take over any form of hand-touch making, it is more about harnessing a new form of intelligence and using this to move creative praxis forwards. In 2019 Paratus People published an article which presented how the Internet of Things (IoT) could close the employment disability gap. Using a vast cloud of ‘IoTT’ data, designers from large media and communication companies e.g. BT and Apple Inc, have designed affordable Assistive Technologies (AT) to aid a disabled population to return to ‘normal’ employment. The cloud-based data supported the use of AT which could be used by over a billion people worldwide who have registered as having some form of physical or sensory ability needs. according to WHO.(

Pacey’s diagram relates technology into various cultural, technical and organizational practices. The triangular region shows how the current value of technology practice is restricted to a technical engineering value, with the greater discourse around cultural and organisational aspects of craft being set within a technological practice. Figure 2

Fourth Industrial Revolution (4IR)

We are moving rapidly to what has been termed by  Klaus Schwab (2016) as the ‘Fourth Industrial Revolution’ (4IR): a way to describe how boundaries are blurred between the physical, digital and biological world. A merging of technologies such as artificial intelligence, robotics, Internet of Things (IOT), digital fabrication, and machine learning are embedded within the physical world. The use of smart technology is becoming ubiquitous in the modern design world.

Machine learning is one form of artificial intelligence, but one that has grown rapidly in recent years. Its basic idea is to take vast volumes of data and find patterns in these data. Once found, these patterns can be used to make predictions, to classify new and unknown data. It allows us to manage the new data world in a way previously impracticable – but a way that is ever more needed as our digital world generates more and more “big data”.

Machine learning provides designers with an opportunity to create more user-centred products. It can personalise the experiences of individuals that use a product and can give insight into the way users interact with our designs. Machine learning also allows us how to predict user behaviour, and so build more robust designs. For example, Jaguar Land Rover is using AI to learn driving habits and preferences to create the world’s first self-learning car. ( Similarly, applications using natural language understanding have enabled people to interact with computers so that they understand the things they say; digital assistants such as Siri and Alexa are good examples.

The Internet of Things (IoT)

The Internet of Things (IoT) is based upon the idea that such smart appliances can collect and communicate data and that these devices to talk to each other.  The use of AI in smart cities can be profound – cameras can recognise people and faces, monitor crowds, and have been used to predict such aspects as parking and traffic management, waste and disposal management. Whilst there are many positive aspects of this level of smart monitoring, this also comes laden with substantial privacy and security concerns. Increasingly, such impacts on human behaviours come with responsibilities that are now central to the design of these systems. This is yet another “Design for X” to be added to the others of reusability, safety, quality, etc. Figure 3

When opening his research facility in Japan (2017), James Dyson explained that “Almost every product can benefit from AI – lighting, purification, cleaning…”.  At Dyson “we combine leading machine learning and AI techniques with an in-depth understanding of user behaviour” ( Figure 4

In more everyday design work, Microsoft PowerPoint uses AI-powered features to streamline graphic work such as suggesting design ideas and visual layouts to presentations. Figure 5

AI in design also opens up many more possibilities to us as designers, allowing us to manage far larger design spaces in the search for good solutions. But it makes us question creativity and our relationship with technology. Designers need to find meaningful ways to use the technology in their work, rather than feeling disempowered by it.

A good example of this exploration of design spaces can be found in the Nutella marketing campaign launched in February 2017 where a single AI algorithm designed 7-million unique labels designs using variations of colours, shapes, polka dots, zigzags and lines. Figure 6 (


This ‘fourth industrial revolution’ brings fundamental changes to the way we operate and disrupts traditional industries. These emerging technologies bring significant positive impacts but clearly pose more philosophical questions too.

In his article  ‘Artificial Intelligences are quickly becoming better artists. And this is just the beginning’, Roey Tzezana (2017) highlights how AI has an advantage over humans in that ‘it is capable of training itself on millions of samples, [to] constantly improve itself’. He reports how AI has composed music performed by the London Symphony Orchestra, won a literary prize and can identify emotions in people and convey these in an abstract painting.  (

Google’s Deep Dream computer programme (2015) uses an AI algorithm to generate psychedelic and abstract dream-like art such as can be seen in the image below. Google has made the code open source so that designers and artists can modify it and create their own versions. Of course, many would argue that such images are derivative and show little genuine creative spirit; merely reproducing fashionable trope. Figure 7

However, Mario Klingemann, a German artist who uses AI in his work, discusses his views on creativity in an interview with Arthur Miller, Guardian (2019), in which he argued that “Humans are not original. We only reinvent, make connections between things we have seen. While humans can only build on what we have learned and what others have done before us, machines can create from scratch”. But, as we have noted above, AI and machine learning, in particular, are best seen as a way of codifying existing examples – they are not really creating from scratch. Instead, we can view them as a means to represent existing practice and apply it to new situations. Figure 8.

How can AI and Machine Learning aid our work as designers?

Automating complex tasks – such as the campaign of visual design as in the Nutella example gives greater freedom to designers to choose from a much wider pool of ideas – generated at speed.  This enables us to evaluate, prototype and test ideas much more effectively. As designers, the flow of data in our connected world blurs the boundaries between the physical and the technological and enables a meaningful partnership with machines. Thus, enabling digitised global collaboration and new variations of an old craft, new formats and creations.

In the article ‘Art by computers: how artificial intelligence will shape the future of design’ Karla Lant surmises that, “The designers of the near future will be even more creative, acting as curators assisted by technology. In the end, AI will enable designers to create forms that would be impossible for a lone human to construct.”(

Figure Sources


2017 [Figure 1] Available at: <> [Accessed 31 October 2020]

2011 [Figure 2] Available at: <> [Accessed 28 October 2020].

2020 [Figure 3] Available at: <> [Accessed 31 October 2020].

2020 [Figure 4] Available at: <> [Accessed 1 November 2020].

2020 [Figure 5] Slide Presentation, Elouise Huxor, MS Powerpoint – AI Powered Features.

2017 [Figure 6] Available at: <> [Accessed 31 October 2020].

2016 [Figure 7] Available at: <> [Accessed 1 November 2020].

2019 [Figure 8] Available at: <> [Accessed 31 October 2020].


















Leave a Reply

Your email address will not be published. Required fields are marked *