Reclaiming Creativity: How Artists Can Still Create Original Art in the Age of Generative AI

As generative AI becomes increasingly embedded in the fabric of contemporary culture, questions about authorship, originality, and artistic ethics loom large. The debate isn’t just academic — it’s visceral, real, and unfolding in art galleries, courtrooms, and online spaces. A recent protest against the UK government’s AI policies, and an open letter demanding that Christie’s cancel an AI-assisted art auction, underscore the growing tension between technology and traditional artistic rights. Yet amid the uproar, a critical point is often missed: generative AI doesn’t have to replace creativity — it can reframe and even deepen it.

In a recent article, Professor Anthony Downey of Birmingham City University confronts this issue head-on. Drawing from his collaboration with renowned artist Trevor Paglen, Downey challenges us not to fear AI as a tool of replication, but to engage with it critically — from within. He argues that by understanding how datasets are constructed, labeled, and manipulated, artists can reclaim agency over AI-generated works. This isn’t about using AI passively; it’s about actively interrogating and subverting the systems that underpin it.

Take for example Holly Herndon and Mat Dryhurst’s work xhairymutantx. By creating custom datasets labeled with manipulated versions of Herndon’s image, they exposed the biases and inconsistencies of AI training models like Midjourney. Likewise, Refik Anadol’s Machine Hallucinations used satellite data to produce endlessly shifting AI-generated landscapes, reminding us of the surreal, hallucinatory nature of machine vision — and the ways it differs fundamentally from human perception.

Paglen’s haunting image Rainbow, created from a dataset he curated and labeled himself, similarly reveals how AI sees the world not as we do, but through warped reflections of its training data. These works are not mere aesthetic experiments — they are investigations into the logic of AI itself.

Downey's central thesis is clear and compelling: the true value of art in the age of AI lies not in competing with machines, but in exposing their inner workings. Rather than dismiss generative AI as a threat to creativity, we must approach it as a complex cultural phenomenon — one that demands transparency, accountability, and critical engagement.

Artists like Paglen, Herndon, and Anadol show us that we can still make original, meaningful art in this new landscape. But doing so requires more than mastery of tools — it demands intellectual rigor, ethical awareness, and a willingness to question the systems we increasingly rely on. It is through the humanities — through art, critique, and reflection — that we can begin to see AI not just as a technology, but as a mirror to our own cultural values and contradictions.

Source: The Conversation