The History of AI

Wu's method

Era 2 · AI is born (1943–1979) › Symbolic AI

Lineage
Symbols and logic
Kind
Theory

Who

Chinese Academy of Sciences

What happened

In 1976 and 1977, the Chinese mathematician Wu Wen-Tsün of the Chinese Academy of Sciences finds a method for proving theorems of elementary geometry mechanically, by translating them into polynomial equations.

Why it mattered

The first practical method of automated theorem proving in geometry: a computer proves hard theorems in seconds. It opens the line of research known in China as the mechanization of mathematics.

The full story

Automated proof with general logical methods, such as resolution, worked poorly in geometry: theorems can be stated in logic, but the search for a proof got lost in the number of possibilities.

Wu, a distinguished topologist, became interested in the problem in the mid-1970s. The method translates the hypotheses and the conclusion of a theorem into polynomial equations among the coordinates of the points. It then arranges the hypothesis equations in triangular form, using Joseph Ritt's theory of characteristic sets, and checks through successive divisions whether the conclusion polynomial vanishes whenever the hypotheses hold.

The method proves theorems that do not depend on the notion of one point lying between two others, and the computers of the day applied it to theorems such as Morley's in little time. Wu set out its principles in an international journal of automated reasoning in 1986.

The work created a school, the mechanization of mathematics, with its own laboratory at the Chinese Academy of Sciences, and the method is also applied to computer vision and robot kinematics.

Sources

How to cite

“Wu's method.” The History of AI, revised November 13, 2026. https://thehistoryofai.com/en/milestone/wu-method

This entry was last reviewed on November 13, 2026.

Open in the interactive timeline