This study investigates the evolving portrayals of women in Chinese state-owned media through a computational analysis of the People’s Daily corpus from 1940s to 2000s. Chal- lenging traditional binary frameworks that simplify portrayals into either masculinized or feminized images, this research applies computational social science methods such as node embedding and topic modeling to identify the dynamic shifts in the official gender discourse system. The empirical findings reveal that, rather than shifting between two idealized states, women’s representations strategically adapted to the state’s changing ideological and political objectives. Three distinct historical periods are identified: the early Mao’s era characterized by rather diverse representations aligning with revolutionary mobiliza- tion; the Cultural Revolution marked by singular, intensified political portrayals; and the post-reform era that witnessed a revival of traditional familial roles along with new represen- tations emphasizing individual economic success and national pride in sports. By reframing women’s portrayals through the lens of gender fluidity and performativity, this research challenges oversimplified masculinization narratives about the Mao’s era and provides new insights into how state power strategically adapt its construction of gender norms.