![]() In recent years machine learning technologies have been advancing rapidly and they are getting more and more accessible as well. These factors enabled artists, designers and just curious individuals to experiment and create with these technologies. There's a multitude of different technologies and usages for machine learning algorithms but in this context I will be talking about image and video GAN’s (A generative adversarial network (GAN) is a machine learning (ML) model in which two neural networks compete with each other to become more accurate in their predictions.). These algorithms generate images or videos based on different inputs such as text, a database of images, video or many more. The images and videos generated by these algorithms all have a common AI aesthetic to them which inspired me to connect it to Anti-design. The classic traits of the anti-design movement can also be recognised as a pattern in modern ML generated imagery. Elena Martinique in her essay defining the roots of the anti-design wrote “Anti-design embraced various exaggerated and expressive qualities to undermine the purely functional value of an object.” if we look at ML generated furniture for instance you can see definite resemblance to the style she mentioned. But these similarities are not only limited to the final products of these but also the processes they create, Dr Marcus Bunyan defined that Modernist designers were often treated as heroic figures, Anti-design tried to challenge this by allowing the owner to participate in the design to some extent this involvement is a core process in ML generating processes especially with openly available accessible algorithms anyone can play the part of the “designer” by using a publicly available pre trained model. Bunyan also wrote that “Anti-Design often suggested that form doesn’t have to be invented, but also recycled.” which can be connected to how most GAN models work, they are trained on a huge database of images for instance if the desired output is to generate images of chairs programers would train the ML model on thousands of images of existing chairs thus “recycling” these chair designs and creating new ones from them using ML tech. I believe that Machine learning technologies are a new tool that inspires and enables a fresh wave of anti-designers of the future. References: https://www.serpentinegalleries.org/art-and-ideas/aesthetics-of-new-ai/ https://creative-ai.org/ https://www.widewalls.ch/magazine/anti-design-italian-movement https://artblart.com/tag/calvin-klein-party/ http://antiaiai.info/ https://www.youtube.com/watch?v=ohmajJTcpNk&ab_channel=MatthiasNiessner https://toretei.com/work_aimobili.html https://www.cadalyst.com/%5Blevel-1-with-primary-path%5D/industrial-design-team-turns-ai-quickly-generate-form-studies-43711 https://affinelayer.com/pixsrv/ David Varhegyi
1 Comment
sarah temple
1/7/2022 12:03:11 am
Excellent and precise response to the manner in which AI democratizes the design process, celebrating the vital act of the user participating in the activity. It reflects debates in law and medicine, for instance in which AI can only 'recycle' former examples with no ability to speculate, express or predict with human capacity. There is a thesis here David - suggesting that AI contribution is limited to visual aesthetics, the look but not the sensory 'feel' or the experience.
Reply
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. Archives
February 2023
Categories |