In today’s rapidly evolving landscape of artificial intelligence (AI), the boundaries between human and machine creativity are being redefined. As generative AI tools continue to infiltrate various fields, particularly the arts, they challenge long-standing ideas of artistic autonomy, control, and authorship. Our latest paper, “Calculated Randomness, Control, and Creation: Artistic Agency in the Age of Artificial Intelligence,” authored by Mariya Dzhimova, with myself as co-author, tackles these crucial issues head-on.
We set out to explore not just the technical capabilities of AI, but also the deeper philosophical implications of human-machine collaboration in artistic creation. We draw comparisons to past art movements, such as Dada and Surrealism, which also questioned the autonomy of the artist by incorporating randomness and rule-based systems into their works.
Click on the image below to download the paper.
in the Age of Artificial Intelligence.
I am really happy to have contributed to this paper, which is part of an ongoing scholarly conversation on AI’s impact on artistic practice (see other publications here). Below, I’ll summarize some of the core ideas and key insights that we defend in the paper.
Key Questions We Address:
- Can AI possess artistic agency, or is it merely a tool in the hands of human creators?
- How does AI-driven art compare to earlier art movements that questioned artistic control and creativity?
- Does AI art represent a new form of distributed agency where both the machine and the human are co-creators?
Key Insights from the Paper:
In the paper, we argue several important points that push forward our understanding of AI’s role in art. Here are the key takeaways:
- AI Art as Co-Creation: We explore the concept of distributed agency, where both the artist and AI shape the creative process. Unlike traditional artistic tools, AI not only executes commands but interacts with the artist in ways that make it a co-creator.
- Historical Continuities: The use of AI in art is not entirely novel. Movements such as Dada and Surrealism also embraced randomness and lack of control to challenge traditional notions of artistic agency. We draw parallels between these movements and contemporary AI art, showing how these historical precedents help contextualize today’s debates.
- AI as a Complex System: AI art functions as an interplay between randomness and control. Unlike rule-based systems in traditional art, AI introduces a layer of unpredictability due to its generative models. This oscillation between calculation and randomness is a central characteristic of AI-driven art.
- Intentional Loss of Control: One of the major shifts with AI art is that artists intentionally delegate parts of the creative process to machines, embracing the unknown and the unexpected. This controlled loss of control allows for a “calculated randomness” that has become an essential part of contemporary AI-driven artistic practice.
- The Dance of Agency: Borrowing from Actor-Network Theory, we describe AI art as a “dance of agency” where human and non-human entities co-shape the final output. The artist initiates a process with AI, but the final outcome is often beyond the artist’s control, leading to an iterative loop of feedback between artist and machine.
- Beyond Anthropocentric Creativity: The integration of AI into the creative process forces us to reconsider our anthropocentric view of art. AI tools act as mirrors or prisms, reflecting the biases and preferences of the artist while also introducing their own generative possibilities. This interaction can lead to the artist discovering new aesthetic directions previously unimagined.
-
The Role of Feedback Loops: We highlight how many artists who work with AI describe the creative process as a continuous feedback loop. AI generates unexpected outputs, prompting the artist to adapt their vision or further refine their prompts. This iterative loop allows for self-reflection and a deep engagement with one’s own artistic practice.
The Significance of AI Art Today
In the paper, we argue that AI art not only challenges our understanding of creativity but also brings new relevance to concepts like hybridity, co-creativity, and distributed agency. While AI may be a tool, it is unlike any artistic tool that has come before. Its ability to surprise, challenge, and mirror human creativity is making the relationship between artist and machine more of a partnership than ever before.
AI art is often framed as a demonstration of collaboration between human and machine—a co-creative process that stages and reveals the distribution of agency. The works produced by contemporary AI artists not only display human ingenuity but also the emergent qualities of artificial systems. As AI continues to develop, we believe these collaborations will only grow more complex, raising new questions about control, authorship, and the boundaries of human creativity.
Final Thoughts
In “Calculated Randomness, Control, and Creation: Artistic Agency in the Age of Artificial Intelligence,” we take a deep dive into the evolving relationship between human and machine in the realm of artistic creation. Our goal is to provoke thought, inspire further research, and contribute meaningfully to the broader conversation about the role of AI in art.
I am deeply satisfied with the work we’ve produced and grateful to have had the opportunity to collaborate on such a relevant and thought-provoking topic.
If you’re interested in these discussions, you can read the full paper here.
