Background <p>Prefrailty is a reversible transitional stage of frailty and is associated with impaired postural stability and increased fall risk. Tai Chi may improve balance-related performance in older adults, but its effects on postural stability and gait biomechanics in prefrail populations remain insufficiently characterized.</p> Methods <p>In this single-blind randomized controlled trial, 68 prefrail older adults defined by the Fried Frailty Phenotype were allocated to a Tai Chi plus health education group or a health education control group. Sixty-four participants completed the study (mean age 71.1 years; 31 women; 33 men). The intervention group performed 24-form Tai Chi three times weekly for 12 weeks, while both groups received health education. Functional postural stability was assessed using the Timed Up-and-Go Test (TUGT), Berg Balance Scale (BBS), and Short Physical Performance Battery (SPPB). Static balance was evaluated using center-of-pressure (COP) measures, and gait biomechanics were assessed using three-dimensional gait analysis. Continuous outcomes were analyzed using baseline-adjusted ANCOVA, with complementary group × time interaction analyses. Exploratory change-score correlations were conducted within the intervention group, with full-sample sensitivity analyses adjusted for group assignment.</p> Results <p>After 12 weeks, the intervention group showed better postural stability than controls, including shorter TUGT time (<i>P</i> &lt; 0.001), higher BBS (FDR <i>q</i> &lt; 0.001) and SPPB scores (FDR <i>q</i> = 0.002), lower COP path length and sway velocity under eyes-open conditions (both FDR <i>q</i> = 0.003), and lower COP path length, sway ellipse area, and sway velocity under eyes-closed conditions (FDR <i>q</i> = 0.014–0.018). Gait analysis showed significant improvements in gait speed (FDR <i>q</i> = 0.033), gait profile score (GPS; FDR <i>q</i> &lt; 0.001), gait deviation index (GDI; FDR <i>q</i> &lt; 0.001), and hip rotation deviation (FDR <i>q</i> = 0.007), whereas other joint-specific kinematic variables were only nominally significant. Group × time analyses supported changes in TUGT, BBS, SPPB, gait speed, GPS, and GDI. Exploratory correlation analyses showed that changes in postural stability were mainly associated with changes in hip kinematics, step length, and selected ankle-related features; however, these associations did not remain significant after FDR correction.</p> Conclusion <p>A 12-week Tai Chi program improved functional performance and static postural control in prefrail older adults and was associated with favorable changes in gait speed and global gait-quality indices. Correlation findings should be interpreted as preliminary mechanistic clues. Larger studies with longer follow-up are needed. </p> <p><i>Trial registratio</i>n Chinese Clinical Trial Registry, ChiCTR2300073905, Registered on [2023-07-25]. <a href="https://www.chictr.org.cn">https://www.chictr.org.cn</a>.</p>