The Yixing Zisha teapot sits neatly at the intersection of art, engineering, and heritage. As such, it makes for an intriguing test case for a technology increasingly taking on creative tasks across almost every field of human endeavour, artificial intelligence, or AI. Research in the International Journal of Systems, Control and Communications describes a computational system that can help AI generate new teapot designs without losing the cultural characteristics that make the craft distinctive.
That distinction matters. Generative AI may produce an attractive image or form, but calling the result design can obscure the human knowledge traditionally involved in making an object. At a time when AI-generated art is attracting widespread criticism for flooding creative spaces with superficially polished but culturally thin material, the researchers are attempting something more demanding. They hope to give the machine constraints that reflect the knowledge behind the tradition.
The researchers combined three methods: grounded theory, which identifies concepts from interview data; the analytic hierarchy process, which assigns relative importance to evaluation criteria; and fuzzy comprehensive evaluation, a mathematical method for handling uncertain or subjective judgements. The resulting system scores the cultural value of AI-generated teapot designs and feeds the results back into the generation process.
The team identifies four levels of cultural characteristics: instinctive, behavioural, reflective and evolutionary. These are derived through expert interviews and qualitative coding, allowing the system to identify departures from desired cultural and functional characteristics and signal adjustments.
In simulations, the system distinguished between designs and showed resistance to disturbances in the data. The wider significance is that AI might be used not simply to generate more heritage-inspired objects but to impose some of the cultural knowledge that makes them meaningful.
He, Y. (2026) ‘An intelligent fuzzy evaluation model for digital heritage design: a cyber-physical systems perspective on closed-loop feedback mechanisms’, Int. J. Systems, Control and Communications, Vol. 17, No. 6, pp.1–16.
News media may use this press release as source material, in whole or in part, provided the content is not materially misrepresented. A link back to the original article is appreciated.