Blind Spots in AI Just Might Help Protect Your Privacy

Researchers have found a potential silver lining in so-called adversarial examples, using it to shield sensitive data from snoops.

from Wired https://ift.tt/2mTfUSp
via