Who asks the questions when AI provides the answers?

During a roundtable discussion with ten participants and the three co-curators on AI and art that took place after a guided tour through the Noorderlicht exhibition Pixel Perceptions, urgent questions were raised by the participants: What is AI doing to creativity? How is it changing the way we experience art? And most importantly, what does this reveal about us?

AI is not new. Although most people think its a recent technology, the first experiments with machine intelligence were conducted as early as the late 1950s. (1) Not sursprisingly, AI has been deployed in the military quickly after that, when the US Department of Defense began training computers to mimic basic human reasoning. In medicine, AI has been analyzing bodyscans for years. (2) But what’s new is the general accessibility to tools like ChatGPT and DALL-E. These technologies often seem to sideline human input, delivering polished results that can appear almost like magic. Is this a tool or just a trick?

Artists are not just using AI to produce new work; they are leveraging it to ask the questions that matter. In our exhibition Pixel Perceptions, the work Breaking the News by Noorderlicht Award winner Franc Archive challenges viewers to reconsider whether they can still trust the news and their own perception. The impact of AI on visual culture, communication, labor, and power is vividly illustrated in the 24-meter-long visual by Kate Crawford and Vladan Joler.

The work Ctrl.Alt.Img. by affect lab questions whether AI amplifies stereotypes, encouraging critical reflection on its societal influence. What does AI challenge in originality and authorship? Boris Eldagsen’s work raises these questions further, probing what prompts were used to create his work. These artworks, unlike technology, doesn’t simply 'solve' problems—it forces us to grapple with them. It translates the abstract into the visceral, making complex issues tangible. Yet art also exposes AI’s limitations. Many AI tools produce flawless, hyper-polished outputs that lack the very thing that defines human creativity: the unexpected. It is imperfection, unpredictability, and risk that elevate art. Do we want art to strive for perfection or value art that challenges and surprises us?

A striking example from the discussion involved one of Rosa Wevers’ students in Fine Arts at Academie Minerva, Willie Prisire, who used AI to compose a piece of music that she then adjusted to use in an artistic performance. However, rather than treating the AI as a substitute or endresult for her creativity, she approached it as a creative collaborator. She then paired the AI’s input with her electric guitar to create an exhilarating and deeply personal performance. This illustrates AI’s potential to augment human creativity, not replace it. But even here, it’s clear: without human insight, choices, and emotion, AI remains an instrument. This example does raise critical concerns: what happens to the value of art if AI takes over part of the creative process? Art isn’t just about the final product—it’s about the intention, the creative process, and the story behind it. Who made it, and why? What societal values and personal stories does it reflect? These are questions that AI, however advanced, cannot answer. Art is a distinctly human endeavor.

Another, perhaps more pressing aspect is AI's environmental cost. (3) These systems consume staggering amounts of energy, and the servers that power them leave a heavy ecological footprint. AI’s reliance on vast data centers, often running 24/7, requires enormous electricity consumption. According to some estimates, training a single advanced AI model can emit as much carbon dioxide as five cars over their lifetimes. All these environmental impacts are expected to escalate considerably, with the global AI energy demand projected to exponentially increase to at least ten times the current level and exceed the annual electricity consumption of a small country like Belgium by 2026. (4) The irony is real: while AI promises innovation and efficiency, its backend operations are energy-hungry and unsustainable. The environmental toll doesn’t stop there: these datacenters rely on extensive water resources to cool their systems, often in regions already facing water scarcity. Moreover, the hardware production—the chips and servers—depend on rare earth minerals. Mining these materials is linked to habitat destruction, labor exploitation, and geopolitical conflicts. While this technology moves faster than our ethical frameworks can keep up, offering possibilities but also profound risks. What kind of future do we want? Will AI be a tool for creativity, used responsibly and wisely? Or will we surrender to convenience?

Art can be a compass in these times, to remind us that technology is not an end but a means to something greater, to what is possible and what is meaningful. AI may be powerful, but our decision is still to decide on its direction. Let’s find that future where AI is intelligent and profoundly human.

-

The event Digital Mirror: curator roundtable was organized by Noorderlicht on Wednesday 18 December 2024.

Footnotes:

1. 1956 Dartmouth Summer Workshop, see https://home.dartmouth.edu/about/artificial-intelligence-ai-coined-dartmouth and https://en.wikipedia.org/wiki/Dartmouth_workshop

2. Mohamed Khalifa & Mona Albadawy. 'AI in diagnostic imaging: Revolutionising accuracy and efficiency.' Computer Methods and Programs in Biomedicine Update, Elsevier, 2024. https://www.sciencedirect.com/science/article/pii/S2666990024000132

3. Shaolei Ren and Adam Wierman. "The Uneven Distribution of AI’s Environmental Impacts," The Big Idea, Harvard Business Review. July 15, 2024. https://hbr.org/2024/07/the-uneven-distribution-of-ais-environmental-impacts

4.  'Energy Consumption in Belgium.' World Data Info. December 2024 https://www.worlddata.info/europe/belgium/energy-consumption.php