Effects of a ChatGPT-generated eccentric training programme on speed, change of direction, agility, and jumping performance in U14 tennis players: A non-randomised controlled study

This non-randomised controlled-trial aimed to examine the effects of a ChatGPT-generated eccentric training (ET) programme, which was subsequently validated by human experts before implementation, on physical fitness in U14 tennis players. Twenty-four youth players were recruited and assigned to either the ET-group (n = 13; 7 females; age = 13.19 ± 0.98 years) or the active control-group (CG; n = 11; 6 females; age = 13.13 ± 0.52 years). The AI-designed ET programme was implemented over 8 weeks, during which the CG continued their regular tennis training without additional ET. Participants were tested on linear-sprint-speed (5- and 10-m sprints), change-of-direction (CoD) speed (505-CoD), vertical-jump (countermovement-jump, CMJ), horizontal-jump (standing-long-jump, SLJ), and drop-jump (20-cm drop-jump, DJ-20) performances. Significant group-by-time interactions were found for 5-m sprint, 505-CoD, and Y-Agility performances (p < 0.05), indicating greater improvements in the ET-group (d = 0.6 to 2.15) compared to CG (d = 0.15 to 0.6). For 10-m sprint and jump tests, no significant group-by-time interactions were reported. Individual response analysis revealed that a higher proportion of ET-group participants (73–100%) improved their performance beyond the smallest-worthwhile-change (SWC0.2) threshold across all tests, compared to the CG (18–63%). The present findings support the effectiveness and practicality of the ChatGPT-designed ET programmes for improving physical fitness in U14 tennis players.