This article is by Choi Hye-ri and read by an artificial voice.
The Seoul International Book Fair is exploring how humans should confront artificial intelligence.
The organizers released the theme for 2026 and the accompanying statement on March 7, crediting novelist Kim Yeon-su and two AI models as co-writers — a choice that has stirred debate.
The Seoul International Book Fair takes place in Seoul every June. This year's theme, "Homo duduri," is a portmanteau of "homo sapien" and duduri, a Korean mythical figure known for confronting fire rather than fleeing from it. Duduri, an early form of dokkaebi, or goblin, also historically refers to a blacksmith.
The statement uses that imagery to frame how humans, especially those in the publishing industry, should respond to rapidly advancing AI.
"Humans have created things with fire ever since the first person approached it without fear," the statement reads. "And AI is no different from fire. The issue lies not with the element itself but with what we forge with it."
But readers on social media were more focused on the names listed beneath the text than on the metaphor. The statement credits Kim, a writer known for the 2007 novel "Whoever You Are, No Matter How Lonely" (translated); Claude Sonnet, an AI model developed by the U.S. startup Anthropic; and Gemini, an AI system developed by Google.
Some readers argued that the choice reflected the theme itself, as a statement about living with AI should involve AI in its creation. But many others reacted with surprise or skepticism, commenting, "I can't believe the book fair entrusted its foreword to AI" and "This is shocking."
According to a planner for the Seoul International Book Fair, the decision to use AI to write the statement was a result of the theme itself.
"This year marks the 10th anniversary of the match between Lee Se-dol and AlphaGo, and it is also a time when the use of AI in publishing has become a major topic," the planner said, referring to the Go match between the renowned player and the AI program developed by DeepMind Technologies.
"At the time, [humans and AI] were seen as rivals. Now, the relationship has shifted toward coexistence. 'Homo duduri' refers to the human who confronts this new 'fire' and questions how to live with it."
Kim, who has experimented with using AI in his writing process, created the statement with AI models trained on 10 classic texts from the digital library Project Gutenberg by exchanging prompts and responses.
"We wanted to explore AI as a theme in-depth, so it made sense to us to write the statement with AI," the planner continued. "[Kim] also suggested documenting the process and making it public."
The fair plans to release the documented process in a special limited-edition publication that will accompany the event.
The statement and its surrounding controversy have also revived a broader debate in the publishing industry regarding how writers should disclose the use of AI in their work — or whether they should even use it in the first place.
Under current Korean laws, authors who publish online or in print do not have to disclose whether they used AI tools in the writing process. The Framework Act on the Development of Artificial Intelligence and the Creation of a Foundation for Trust, which took effect on Jan. 22, does not explicitly apply to published works.
Lee Yong-hae, a lawyer who serves as advisory counsel for the International Association for Artificial Intelligence and Ethics, said the use of AI in published writing remains largely unregulated.
"It's difficult to interpret the current AI framework law as applying to published works," Lee said. "When a human author uses AI to write, whether they disclose that essentially depends on the author's conscience. In that sense, it remains a legal gray area."
But some publishers say that the issue will eventually require strict standards.
A publishing house chief who asked to remain anonymous said that the rapid development o...