Runner's AI coaching attempt fails

A 37-year-old man from an urban area tried to use ChatGPT as his running coach to prepare for a half-marathon in 1 hour and 45 minutes. However, he did not achieve his goal during the Lyon half-marathon on October 4, 2026.

Training plan was not the sole issue

The runner attributed his failure to the AI's training plan, but a review with professional running coach Thibault Vallé showed that the plan was mostly sufficient and coherent. The sessions were not inherently unsuitable for the task.

AI's limitations in training

Thibault identified several issues with the AI's approach, including its lack of adaptability and the repetitive nature of the training content. The AI also focused more on positive feedback than on addressing the runner's physical limitations or warning signs.

AI training lacks personalization

The AI provided a structured plan with three weekly sessions but failed to adjust to the runner's specific needs or context. The runner admitted he ignored his body's signals, such as feeling heavy in his legs during training.

Human coaching remains essential

Thibault emphasized the value of human interaction, adjustments, and variety in training. He suggested that AI can support coaching but cannot replace the nuanced understanding of an individual's physical condition.

Runner acknowledges personal responsibility

The runner concluded that the issue was not solely the AI's fault but also his own reliance on it and lack of self-awareness. He recognized the need for a more balanced approach to training.