By Tom Simonite
“Are you in a precarious situation? … You sound like you can’t talk.”
Karah Preiss’ cousin Leslie accused her of being sleepy and distracted and eventually hung up, but didn’t guess the truth. Preiss had placed the call using a software clone of her voice made to demonstrate artificial intelligence’s ability to deceive.
Preiss relates her family experiment in the fifth installment of the Sleepwalkers podcast, a guide to the recent boom in artificial intelligence. The episode examines how AI technology is reshaping perceptions of reality in phone pranks, on Facebook, in Hollywood, and in politics. Hosts Oz Woloshyn and Karah Preiss talk to the experts fighting against deep fakes, as well as those using them for good.
In this episode: Nathaniel Gleicher of Facebook, John Micklethwait of Bloomberg News, Jose Sotelo of Lyrebird, Danielle Citron of Maryland Carey Law, Hany Farid of UC Berkeley, and David Kirkpatrick of Techonomy.
Hany Farid is a professor in the UC Berkeley School of Information and EECS. He specializes in digital forensics.