Ars Technicast special edition, part 3: Putting AI to work defending your stuff

Increase The Size Of / Artist’s impression of adversarial AI being adversarial.
Grassetto/ Getty Images

reader remarks

with 8 posters getting involved

In the 3rd and final setup of our podcast miniseries on expert system, produced in association with Darktrace, we look into the world of AI battling AI– or what scientists refer to as “adversarial AI.”

Click on this link for a records and click on this link for an MP3 direct download.

Adversarial expert system can take lots of types– as a tool for hacking through AI- powered security of other systems, for instance, or tricking another algorithm with input that triggers a particular, phonyresult Ars editors Sean Gallagher and Lee Hutchinson spoke to the leader of the winning team from the 2016 Defense Advanced Research study Projects Firm’s Cyber Grand Obstacle, ForAllSecure CEO David Brumley, about improvements in AI- driven hacking. Lujo Bauer, a teacher of electrical and computer engineering at the Institute for Software Application Research Study at Carnegie Mellon, signed up with Lee and Sean to speak about his research study into methods to usage AI to defeat technologies such as facial acknowledgment. And Max Heinemeyer, director of danger searching at Darktrace, talked about research study currently being done into how to stop AI- driven attacks on computer networks.

This special edition of the Ars Technicast podcast can be accessed in the following locations:

iTunes: (May take numerous hours after publication to appear.)




Leave a Reply