Skip to the main content

Bracing Medical AI Systems for Attacks

There’s new advice on how to handle tampering that fools algorithms and enables healthcare fraud

Illustration of an AI doctor examining medical images

Last June, a team at Harvard Medical School and MIT showed that it’s pretty darn easy to fool an artificial intelligence system analyzing medical images. 

The medical industry and AI systems can prepare for adversarial attacks with thoughtful approaches.

Jonathan Zittrain comments on planning smart. “We shouldn't rush to anticipate every terrible thing that can happen with a new technology, but rather release and iterate as we learn.” Trying to anticipate and prevent all possible adversarial attacks on a medical AI system could cripple rollout, delaying the good a system might do, such as diagnosing patients in rural areas who lack access to disease specialists.

Read more at Spectrum...

You might also like

Projects & Tools 01


AI: Transparency and Explainability

There are many ways to hold AI systems accountable. We focus on issues related to obtaining human-intelligible and human-actionable information.