An intelligent health coach that misunderstands movement


A start-up is developing an AI-powered health app designed to help users increase their physical activity and thereby improve well-being in the long term. The app analyzes movement data from wearables and suggests personalized training plans and break times.

At first, many users report motivating and helpful support in everyday life. But after some time, unexpected problems arise: Some users report symptoms of overload, while others are classified as "at risk" despite low activity. The AI seems to partially misinterpret movement patterns, leading to inappropriate recommendations.

The development team faces the challenge of understanding the technical and data-related causes of these misinterpretations and assessing the impact on physical health, trust in digital health coaches, as well as the ethical requirements for such systems.


Question:
Which factors can cause an AI-powered health coach to misinterpret movement data, and how do these misjudgments affect users’ physical health, user trust, and the requirements for the development and deployment of digital health applications?

Solution follows tomorrow.