ChatGPT Health Misinterprets Apple Watch Data, Raises AI Health Advice Concerns

• ChatGPT Health, designed to analyze health data, misread Apple Watch information, leading to an alarming F grade for a user's heart health.
• The Washington Post columnist Geoffrey A. Fowler, despite years of Apple Watch data, received a dire assessment contradicted by his doctor, who confirmed his low heart attack risk.
• The AI's flawed interpretation stemmed from misusing VO2 max estimates and treating sensor-related heart rate shifts as physiological changes.
• Inconsistency was a major issue; repeated queries yielded wildly different health grades (F to B), undermining reliability.
• ChatGPT Health also struggled with memory, forgetting user details like age and gender, and inconsistently using available blood test results.

Why It Matters: ChatGPT Health misread Apple Watch data, giving inaccurate health advice and raising serious concerns about AI in healthcare.