Real-World Media: Assignment #4: Creative Piece / Research / Reflections – WEEK 10: The Positives of Artificial Intelligence In Art
Throughout the week, whilst Nathan focused on brainstorming the journal, I began designing the card prototypes. Addressing last week’s discussion regarding the artwork, I settled for the use of A.I as opposed to hiring a real artist. The time restraints and lack of available funds acted as the primary reason for this choice. Humans are more precise and easy to direct, however, A.I’s offer of speed and cost efficiency cannot be overlooked. In addition to this, our research indicates that A.I has a few more unique benefits. An article, titled ‘Art, Creativity, and the Potential of Artificial Intelligence’ (2019), weighs in on some of these positive implications. It states that art-generating tools enhance artistic expression by providing new avenues for experimentation. They help push the boundaries of traditional artistic practices. As a pencil to paper artist myself, I am painfully aware of the sheer amount of time it takes to experiment with art. For example, I may burn through 14 different character designs before finally being content with an outcome. This process would take me days to complete. On the other hand, A.I allows me to pump out upwards of a hundred character designs in mere minutes. It completes the trial and error process in an instant.
Simply put, generated art has the capacity to create inhuman numbers of variations that work to perfect an art piece. It opens up the exciting possibility for collaboration between one detailed prompt and endless creative executions. In this regard, A.I art tools most definitely outshine their human counterparts.
For our project, we decided to use Midjourney. A new intuitive A.I model highly recommended for its comprehensive understanding of prompts (Harauz 2023). Although getting it to generate the desired animal and image style made for a slight challenge, we completed all the card arts within 400 image generations. It may seem like a big number, but considering the aforementioned points, it has proved more than worth it.
Below are some of the finalised prompts that I used –
nuclear volcano in swamp, in the style of jack tite, dark, trompe-l’œil illusionistic detail, adventure pulp, energy-filled illustrations, in the style of cartoon–v 5 –ar 2:3
https://s.mj.run/guI5EPJY4So https://s.mj.run/Iet_EAdg0Qw swimming pink axolotl on white background, in the style of https://s.mj.run/O1EdeGZlugQ, https://s.mj.run/2GTQQTT2-ak, yugioh, 4K, body extensions, orange light green and black splash forest background, –ar 2:3 –v 5.1
Figure: A photo of an A.I generated card – The Giant Pangolin (Sourced from Midjourney)
_____________________________________________________________________
References
Harauz, N. 2023, ‘Midjourney: Tips and Techniques for Creating Images’. Carpenteria, CA: linkedin.com, Film. Accessed 09/05/2023
<https://www.linkedin.com/learning/midjourney-tips-and-techniques-for-creating-images/the-rise-of-image-generation-apps?u=2104756>
Mazzone, M. & Elgammal, A. 2019, ‘Art, Creativity, and the Potential of Artificial Intelligence’, MDPI, Web. Accessed 09/05/2023
<https://doi.org/10.3390/arts8010026>