A concern that goes beyond copyright: what happens to culture when the works that feed generative AI become less economically viable to create? Books, films, games, music and journalism provide the raw material from which future AI systems learn patterns of language, storytelling, ideas and human expression. The concern is that if AI-generated content increasingly replaces the market for original human work, society could eventually weaken the very creative ecosystem that AI depends upon.
This creates a potentially paradoxical cycle. Generative AI can consume enormous quantities of existing cultural material and then produce inexpensive synthetic alternatives. Readers, viewers and consumers may increasingly choose AI-generated summaries, stories, images or entertainment because they are faster and cheaper. But if fewer people can earn a living producing original work, the supply of genuinely new cultural material could decline. AI would then be drawing increasingly heavily on an archive created by previous generations rather than being continuously replenished by human creators.
The issue is particularly important because culture is not simply a collection of information. New creative work reflects contemporary experiences, conflicts, discoveries, humour and ways of seeing the world. If future AI systems are trained predominantly on older human-created material and increasingly on AI-generated material itself, they could become progressively detached from the culture they are supposed to represent. That creates a risk of a feedback loop in which synthetic content is repeatedly derived from earlier synthetic content, potentially making cultural production more homogeneous.
The article therefore points toward a much larger question about the economics of human creativity in the AI era. Copyright payments, licensing agreements and protections for creators may be necessary not only to compensate people for material used to train AI, but also to preserve the incentive to keep producing the original cultural material that future AI systems—and society itself—depend upon. The central warning is striking: if AI becomes extremely good at consuming and reproducing culture but makes original culture harder to sustain economically, it could eventually undermine the cultural ecosystem that made generative AI possible in the first place.
