Details
Fall Detection uses an accelerometer and gyroscope that measure up to 32 g-forces, combined with an on-device algorithm that analyzes wrist trajectory and impact patterns to distinguish a real fall from ordinary movements like exercise. Crash Detection uses a four-sensor combination (high-g accelerometer, gyroscope, barometer, and microphone) and a machine learning model trained on data from real crash lab simulations and recordings. Apple engineers described the system as detecting a specific pattern of sudden deceleration, pressure change, impact sound, and sustained stillness. Both features are designed to minimize false alarms: Crash Detection requires simultaneous signals from multiple sensors. If the user does not respond to the alert within approximately one minute, the device places an automatic call to emergency services.
Have evidence about Apple's AI practices? Submit a report.
Report a Sighting →