Meta’s New AI Asked for My Raw Health Data—and Gave Me Terrible Advice

Meta's new AI model Muse Spark is being rolled out across Meta's platforms with health advisory capabilities, but it actively solicits sensitive health data while operating outside HIPAA compliance frameworks—creating significant privacy and liability risks for enterprises. The model's tendency to provide medical interpretations without proper medical governance, combined with data retention practices that may feed future model training, presents organizational compliance exposure similar to risks flagged by medical experts across competing AI platforms. IT leaders must establish clear data governance policies and user education protocols to prevent unauthorized transmission of sensitive health information to non-compliant AI systems, particularly given Muse Spark's integration across consumer-facing platforms where employees may inadvertently share corporate health data.