FST.ai 2.0: An Explainable AI Ecosystem for Fair, Fast, and Inclusive Decision-Making in Olympic and Paralympic Taekwondo

AI can reduce bias in sports officiating: FST.ai 2.0 includes fairness monitoring modules designed to track and reduce inconsistent decision-making by referees, addressing a longstanding challenge in combat sports where subjective judgment calls can disadvantage certain athletes or countries.

Transparency builds trust in AI systems: The system achieved 93% referee trust through explainability overlays that show why the AI made specific decisions in real time, demonstrating that when officials understand AI reasoning rather than just receiving recommendations, they're more likely to accept and rely on the technology.

AI can accelerate decision-making without sacrificing accuracy: FST.ai 2.0 reduced decision review time by 85%, meaning referees can make faster calls during competitions while maintaining fairness—a critical benefit for athletes and the flow of competition.

Uncertainty matters in high-stakes decisions: The system uses "epistemic uncertainty modeling" to flag situations where the AI is less confident, preventing overconfident recommendations in ambiguous moments and ensuring humans remain in control of final calls in close competitions.

Inclusive AI requires specialized design: The ecosystem includes dedicated modules for Para-Taekwondo classification and athlete assessment, showing that fair AI systems must be intentionally designed for diverse populations rather than assuming one-size-fits-all solutions work across all athletes and abilities.