AI Medical Systems Vulnerable to Privacy Attacks; North Korean Hackers Compromise AI Framework

Thumbnail Image

The information displayed in the AIM should not be reported as representing the official views of the OECD or of its member countries.

Researchers warn that AI medical systems are susceptible to membership inference attacks, risking patient privacy by revealing sensitive health data. Separately, North Korean hackers infiltrated the Mastra AI application framework, injecting malicious code into over 140 NPM packages, compromising software security and user trust.[AI generated]

Why's our monitor labelling this an incident or hazard?

The article explicitly involves AI systems used in healthcare and highlights a security weakness that could lead to privacy breaches and harm to individuals' rights and privacy. Although no actual harm is reported as having occurred yet, the described attack method plausibly could lead to significant harm if exploited. Therefore, this event constitutes an AI Hazard, as it outlines a credible risk of harm stemming from the use of AI in medical contexts.[AI generated]
AI principles
Privacy & data governanceRobustness & digital security

Industries
Healthcare, drugs, and biotechnologyDigital security

Affected stakeholders
Consumers

Harm types
Human or fundamental rightsReputational


Articles about this incident or hazard