Medical AI could compromise your privacy in disturbing new way, experts warn (opens original article in a new tab)
Experts warn that medical AI could undermine privacy by allowing hackers to infer if a patient's data was used in training models, potentially exposing sensitive health information.
- Medical AI systems may compromise patient privacy through privacy attacks
- Malicious users could determine if a patient's data was used to train AI models
- Membership inference attacks can reveal sensitive medical information with high success rates
Conversation
No comments yet
Threaded discussion is coming next — this is where the community conversation about this story will live.