At Johns Hopkins, Shadow AI Use Linked to Data Breach
OVERVIEWJohns Hopkins experienced a data breach connected to unauthorized ("shadow") AI tool use, highlighting the security risks when clinicians deploy AI systems outside official governance and IT oversight channels. The incident illustrates a critical healthcare IT challenge: clinicians increasingly adopt AI tools (ChatGPT, Claude, specialized apps) without formal institutional vetting or security integration. These shadow AI systems operate outside data protection protocols, creating vulnerability to breaches and compliance violations. The Johns Hopkins case demonstrates that healthcare security frameworks must address not just authorized tools but also the emerging ecosystem of shadow AI. Governance strategies should include: AI tool discovery and inventory; security assessment protocols; clinician training on approved alternatives; and enforcement mechanisms that don't completely block innovation. ---
Subscribe to read full wiki analysis
Full wiki content — including source provenance, contradiction tracking, related pages, and analyst notes — is available to Subscriber members.
View Plans & SubscribeNew users get a 7-day trial — no credit card required.