If you can write it, you can do it: the rise of AI-generated content
In 2019 at the preview for Art Basel Miami Beach a banana taped to the wall (Figure 1) was sold as a piece of art at the cool price of 120.000 dollars. Subsequent versions from the same artist saw an increase in their price [1]. In August 2022, Jason M. Allen won the first prize (300 dollars) at a Fine Arts Competition in Colorado by submitting a digital image (Figure 2) [2]. Both recent art episodes provoked turmoil. The first, for obvious reasons. The second, because the winning image was generated entirely by an artificial intelligence (AI) program.
The debate around what’s art and what’s not has puzzled numerous cultures and societies. Far from being solved, the appearance of the ‘AI artist’ in the scene has already rocked the boat.
Figure 1: Maurizio Cattelan’s Comedian, for sale from Perrotin at Art Basel Miami Beach. Source: Sarah Cascone [1].
Figure 2*: Jason M. Allen’s piece “Théâtre D’opéra Spatial” which he created with AI image generator Midjourney. Source: Rachel Metz [2].
One of the most impressive examples of AI-generated content is the development of machine learning algorithms that can create realistic images and videos. These algorithms, known as generative adversarial networks (GANs), work by training a machine learning model on a large dataset of images or videos. The model then uses this training to generate new content that is similar to the examples in the dataset.
Figure 3*. Source: Ido Beeri [6].
AI is also being used to generate text content, such as news articles and social media posts. In some cases, these AI-generated texts are indistinguishable from those written by humans. This has led to concerns about the potential for AI to replace human writers and journalists in the future. However, it is also worth noting that AI-generated text has the potential to assist human writers in the creative process and to enhance the efficiency of certain tasks, such as data analysis and research.
Figure 4*. Source: Keith Burgess [7].
In light of these concerns, it is important that the development and use of AI-generated content be carefully regulated and monitored. This may require the development of new ethical guidelines and the creation of oversight bodies to ensure that AI is used responsibly and ethically.
Overall, the ability of AI to generate content is an impressive and potentially transformative development. However, it is important to approach this technology with caution and to consider the potential risks and ethical implications of its use. By carefully managing the development and use of AI-generated content, we can ensure that it is used for the benefit of society and not to the detriment of individuals or groups.
Figure 5*. Source: James Gill [8].
The above paragraphs in italics have been generated by an AI algorithm, called ChatGPT [3], from an input of a few words:
‘Write a 700 words essay about AI now being able to generate content (e.g., images, text, video).”
The information written in those paragraphs has not been fact-checked, therefore could be inaccurate. Whereas the images marked with an asterisk have been generated by another AI, DALL-E 2 [4], always from a text string.
Having tried the AI to generate content for this and other exercises, the best summary of what AI like ChatGPT constitutes is: “a plausible idiot” [5]:
“It gets just enough right, saying just enough words, to sound plausible and authoritative to anyone who doesn’t know the subject matter well. But it also gets enough wrong that you cannot rely on its accuracy, and if it is talking about a subject you know well it is sometimes laughable how wrong it is.”
Clearly, this technology is not able to entirely replace human creativity (yet). Surely, AI has the potential to support and complement the work of humans by providing them with new tools and resources to create content at an unprecedented speed and ease. Humans bring a unique and valuable perspective that cannot be replicated, however AI has already changed the industry of content creation as we used to know it.
[1] Cascone (2019). Maurizio Cattelan Is Taping Bananas to a Wall at Art Basel Miami Beach and Selling Them for $120,000 Each. https://news.artnet.com/market/maurizio-cattelan-banana-art-basel-miami-beach-1722516. Accessed on 17/12/2022. [2] Metz (2022). AI won an art contest, and artists are furious. https://edition.cnn.com/2022/09/03/tech/ai-art-fair-winner-controversy/index.html. Accessed on 17/12/2022. [3] codingdave. https://news.ycombinator.com/item?id=33863563. Accessed on 17/12/2022. [4] https://chat.openai.com/chat. Accessed on 17/12/2022. [5] https://openai.com/dall-e-2/. Accessed on 17/12/2022. [6] Ido Beeri. Generated using DALL-E 2 from the prompt: “A children’s book with beautiful elephant trunks made of hyperrealistic impossible tesseracts, about the boy who wanted to be a cauliflower. Written by Joan of Arc and Illustrated by Albert Einstein himself. In shades of red, embossed in papers made of iron. Extremely detailed photography with all elements, f/1.8.” https://www.facebook.com/photo?fbid=5749060345181941&set=pcb.686953656332661. Accessed on 17/12/2022. [7] Keith Burgess. Generated using DALL-E 2 from the prompt: “Salticidae Celebrating Saturnalia.” https://www.facebook.com/groups/dalle2.art/permalink/684302463264447/. Accessed on 17/12/2022. [8] James Gill. Generated using DALL-E 2 from the prompt: “Superman saves Christmas, painted by William-Adolphe Bouguereau.” https://www.facebook.com/photo.php?fbid=10159273294882333&set=p.10159273294882333&type=3. Accessed on 17/12/2022.
Add a Comment