Apple's Face ID facial recognition technology on iPhone X fails to recognize users in the morning due to facial swelling and changes in eye appearance, forcing users to enter passcodes instead.
Apple's Face ID facial recognition system on the iPhone X experiences recognition failures when users attempt to unlock their phones in the morning. The system, which uses a TrueDepth camera to project 30,000 invisible dots onto faces to create depth maps and mathematical representations, struggles to identify users whose faces show morning puffiness or swollen eyes. Several users, including journalist Connie Wang from Refinery29, reported that Face ID consistently fails to recognize them during the first 30 minutes after waking up, regardless of lighting conditions or whether they wear glasses. The issue appears particularly related to eye appearance, as users with allergies or those who have been crying also experience recognition failures. Security expert Marc Rogers noted that Face ID assigns higher value to certain facial parts and requires open eyes to function properly. While Face ID can recognize users through various accessories like sunglasses, makeup, and even shaving foam, the subtle facial changes that occur during sleep prove challenging for the algorithm. Users affected by this issue must resort to entering their passcode to unlock their phones, which some describe as a 'humiliating neg' that makes them feel judged by their device. Apple has not provided specific solutions for this morning recognition issue beyond general troubleshooting tips.
Domain classification, causal taxonomy, severity scores, and national security assessments were LLM-classified and may contain errors.
AI systems that fail to perform reliably or effectively under varying conditions, exposing them to errors and failures that can have significant consequences, especially in critical applications or areas that require moral reasoning.
AI system
Due to a decision or action made by an AI system
Unintentional
Due to an unexpected outcome from pursuing a goal
Post-deployment
Occurring after the AI model has been trained and deployed