The Surprising Economics of AI in Filmmaking: A Costly Future?

You know, when we talk about artificial intelligence in creative fields, especially filmmaking, the conversation often quickly turns to efficiency and, perhaps most importantly, cost-cutting. The general consensus, it seems, is that AI tools are here to streamline workflows, reduce manual labor, and ultimately make production more affordable. But hold on a minute, because some leading experts are now suggesting a rather counter-intuitive scenario, one that could genuinely reshape our understanding of AI's economic impact on cinema.

Indeed, a recent gathering at the Taiwan Creative Content Fest (TCCF) in Taipei on November 8, 2026, saw a truly fascinating, if somewhat unsettling, warning emerge. Two prominent Korean AI pioneers, Kim Kwang-jib and Lee Sang-wook, stepped forward with a sobering prediction: producing a feature film using AI might just end up being more expensive than even the fabled 35mm film era. Quite a claim, wouldn't you say? Especially given how much we romanticize the perceived expense of traditional celluloid production.

Their primary concerns revolve around what they've termed 'runaway token fees' and the sheer complexity of 'piecemeal revisions.' Now, what exactly does that mean for a budding filmmaker or a seasoned producer? Well, imagine every single adjustment, every tweak, every creative decision, no matter how minor, incurring a micro-charge – a 'token fee' – that accumulates rapidly. Then, consider the iterative nature of filmmaking, where scenes are endlessly refined, characters are adjusted, and narratives are subtly reshaped over and over again. Each one of those 'piecemeal revisions' would, presumably, rack up additional costs, pushing the overall budget surprisingly high.

It’s a fascinating paradox, really. While AI has undeniably proven its worth in making shorter-form content, like commercials, significantly cheaper and faster to produce, this cost-saving magic might not, in fact, translate to the grand scale of a feature film. The intricate storytelling, the nuanced character development, and the sheer volume of assets and scenes required for a full-length movie present a vastly different challenge. The constant back-and-forth, the creative explorations unique to longer narratives, could make AI-driven production a financial quagmire rather than a streamlined highway.

So, as the industry hurtles towards an AI-integrated future, this warning from Kim Kwang-jib and Lee Sang-wook serves as a crucial reminder. It's not just about the technology itself, but the economic models and billing structures that accompany it. Perhaps it's time for the industry to carefully consider how these 'token fees' are structured and how revisions are handled, lest the promise of AI efficiency turns into an unexpected budget nightmare, making the 'golden era' of 35mm look comparatively economical.