Associate Professor of Art Andrew O’Brien will incorporate artificial intelligence prominently into his new fall 2024 Art 4900r course, “Special Topics: Art and Artificial Intelligence.”
O’Brien has identified course learning objectives that center on current or anticipated use of AI in creating works of art and student understanding of the foundational technology and components of AI platforms; of using AI to improve research and ideation; and ethical implications associated with AI.
Readings, discussions and writing assignments for students will address technology’s impact on contemporary art and its role in their own evolving studio practice. In collaborative processes with AI systems, students will demonstrate their understanding of the interactive possibilities between human artists and artificial intelligence, O’Brien said.
Students will create artworks for the course that reflect a nuanced understanding of the intersections of creative concepts originating from established analog and digital art processes, and those developed using AI, O’Brien said. Course work will require addressing technical and conceptual challenges of integrating AI technologies into artistic practice.
“An important facet of the course will be to situate AI within historical uses of and debates surrounding new technology in modern and contemporary art practice,” O’Brien said. For example, students will read the pivotal 1935 essay, “The Work of Art in the Age of Mechanical Reproduction,” by Walter Benjamin, a philosopher, cultural critic and media theorist whose work explored the impacts of new technology—specifically photography and the cinema—on art and culture in the early 20th century. Benjamin’s perspective will be considered in the context of current views on the role of AI in art.
Students will consider Benjamin’s notion of authenticity in art in making use of generative AI tools to create two types of images, O’Brien said.
“The first will be an image of a memory created by prompting a text-to-image AI tool such as Open AI’s Dall-E or Stable Diffusion, for which they already have an existing photograph,” he said. “The second image will be a similar prompt process but for a memory that has no photographic record. Students will then reflect on how these newly created images relate to the original photograph as well as the memory that has no existing visual record.”
Another assignment will direct students to engage with new large language model AI tools such as OpenAI’s ChatGPT, GitHub’s CoPilot and Google’s Bard and the platforms’ ability to quickly generate computer code through natural language prompts or queries.
“This allows those with very little coding experience to develop websites and a range of interactive programs,” O’Brien said. “This assignment will ask students to develop a web-based artwork using one of these large language model AI tools. In the process, students will learn about the structure of websites by ‘reverse-engineering’ the code generated through the AI tool.”
O’Brien notes that many early examples of web-based art used simple HTML websites and GIF animations and said his students will be encouraged to consider working with the web as a potential tool for creative expression not constrained by typical uses of the internet for news, shopping or social media.
The project is intended to highlight the concept of “collaborative AI,” in which engagement between human user and generative AI tool enable development and evolution of an idea through a series of prompts that build upon one another. O’Brien’s proposal to develop a new AI-centered course is supported by a $1,500 stipend from College of Arts and Sciences Dean Pam Riggs-Gelasco.