Inside the Fight Against Non-Consensual Deepfake Porn

**Kara Swisher** (0:00)
Hey folks, a word of warning before we begin. Today's episode discusses sexual violence, including rape, graphic sexual content, and suicide. So you might want to keep this in mind when you decide if and where to listen.
Hi, everyone from New York Magazine and the Vox Media Podcast Network. This is On with Kara Swisher, and I'm Kara Swisher. My guests today are attorney Carrie Goldberg, tech journalist and CEO of Mostly Human Media, Laurie Segall, and VS. Subrahmanian, the Walter P. Murphy Professor of Computer Science at Northwestern University. We're talking about non-consensual deepfake pornography, or AI-generated explicit videos and images that use real people's likenesses without their consent.
Celebrities were often the target of early pornographic deepfakes. Now, according to one survey, one in 18s personally knows someone who has been targeted. Nearly all deepfake porn depicts women. Despite laws and moderation efforts, pornographic deepfake websites and so-called Nudify apps continue to multiply.
A recent analysis from Wired found that Elon Musk's Grok chatbot is still being used to generate and host non-consensual deepfakes. I am not surprised in the least. Carrie Goldberg is a victim's rights attorney. Her law firm specializes in fighting internet abuse and revenge porn. Journalist Laurie Segall recently released Searching for Mr. Deepfakes, an investigative docu-series created for TikTok in partnership with Paris Hilton. It follows an investigation to unmask the anonymous operator behind one of the internet's most notorious deepfake porn platforms.
VS. Subrahmanian is a top computer scientist and deepfake expert. He led the team that developed the global online deepfake detection system, a free platform to help journalists detect deepfakes. I think it's critically important to talk about this issue. When I started my career, pornography was at the heart of the problems with the early internet, and the uses of it and the abuse of it was built into the system at the start. A lot of people were aware of it then and have had a very hard time dealing with it, even as technology has gotten better and better at being more and more abusive. So it's critical that we keep talking about it, we criminalize this, and we start to hold platforms, big platforms, responsible for the stuff that they distribute. Our expert question today comes from Wall Street Journal tech reporter, Georgia Wells. This is an important conversation and one we will keep having, so stick around. One more thing before we get started, if you're going to be in Washington, DC on July 16th, please join us live at the Johns Hopkins University Bloomberg Center for a taping of On. I'll be talking to Gina Raimondo, Commerce Secretary under President Biden and former Rhode Island Governor, about what AI means for the American workforce. As a special bonus before that conversation, I'll be speaking with Johns Hopkins University President, Ron Daniels and the University of Notre Dame President, Father Robert Dowd about how universities are approaching AI and workforce issues. You can get tickets and learn more at voxmedia.com/karaswisherlive.

**SPEAKER_3** (3:25)
This monday.com ad was created by a team of people and AI agents. The agents wrote the copy and managed the timelines, while our human creative director made sure it all made sense. Easy. Create your own AI agent today on monday.com.

**SPEAKER_4** (3:41)
A lot of companies are investing in AI, but most only have a small group actually using it, and an even smaller group who are really seeing the value. Superhuman fixes that. From the makers of Grammarly, Superhuman AI lives in every tool your team already uses, for instant help with drafts, summaries, and more. So habits and proficiency form naturally. When every person on your team works at their best, that's when your AI investment starts to compound. See what Superhuman can do at superhuman.com.

**Ryan Reynolds** (4:12)
Hey, it's Ryan Reynolds here for Mint Mobile. Now, I was looking for fun ways to tell you that Mint's offer of unlimited premium wireless for $15 a month is back. So, I thought it would be fun if we made $15 bills. But it turns out, that's very illegal.
So, there goes my big idea for the commercial. Give it a try at mintmobile.com/switch.

**SPEAKER_6** (4:33)
Up from payment of $45 for three months, $90 for six months or $180 for 12 month plan required, $15 per month equivalent to taxes and fees extra. Initial plan term only greater than 50 gigabytes to be slow when network is busy. See terms.

**Kara Swisher** (4:46)
Carrie, Laurie and VS., thanks for coming on On.

**Laurie Segall** (4:49)
Good to be here.

**V.S. Subrahmanian** (4:50)
Thrilled to be here.

**Kara Swisher** (4:52)
Hello, hello.
So, I talked to experts back in January about the proliferation of sexualized non-consensual AI deepfakes, including images of children on Elon Musk X that were generated by its Grok chatbot. Obviously, AI created deepfake porn, hasn't gone away on Grok or elsewhere, and the technology continues to improve. I'd like each of you to explain the current scope of the problem and how pervasive it is and who it's affecting and how technically advanced it is. We're going to get more into detail. So, go high level here. Carrie, we'll start with you, then Laurie, then VS.