Researchers suggest medical AI systems could be vulnerable to adversarial attacks

A small team of medical researchers from Harvard University and MIT has published a Policy Forum piece in the journal Science suggesting that future medical AI systems could be vulnerable to adversarial attacks. They point out that prior research has shown that virtually all AI systems are vulnerable in some way to such attacks.

from News on Artificial Intelligence and Machine Learning https://ift.tt/2Wg6A6W
SHARE
    Blogger Comment
    Facebook Comment

0 comments:

Post a Comment