meta_pixel
Tapesearch Logo
Log in
Science Quickly

The Internet Is Full of Deepfakes, and the Sky Is Full of Trash

Science Quickly

Scientific American

Science

4.2639 Ratings

🗓️ 13 May 2024

⏱️ 8 minutes

🧾️ Download transcript

Summary

AI-generated images of Katy Perry at Monday’s Met Gala looked so realistic they even duped her mom. And it just so happens that ChatGPT developer OpenAI released a new tool to detect fake images generated by DALL-E—the very next day.   Join Scientific American, Springer Nature and Nature Portfolio in Washington, D.C. on May 17 for Science on the Hill. Register now! Email us at [email protected] if you have any questions, comments or ideas for stories we should cover! Discover something new everyday: subscribe to Scientific American and sign up for our daily newsletter.  Science Quickly is produced by Rachel Feltman, Kelso Harper, Madison Goldberg and Jeff DelViscio. This episode was hosted by Rachel Feltman. Our show is edited by Elah Feder, Alexa Lim, Madison Goldberg and Anaissa Ruiz Tejada, with fact-checking by Shayna Posses and Aaron Shattuck. The theme music was composed by Dominic Smith. Learn more about your ad choices. Visit megaphone.fm/adchoices

Transcript

Click on a timestamp to play from that location

0:00.0

Understanding the human body is a team effort. That's where the Yachtel group comes in.

0:05.8

Researchers at Yachtolt have been delving into the secrets of probiotics for 90 years.

0:11.0

Yachtold also partners with nature portfolio to advance gut microbiome science through the global grants for gut health, an investigator-led research program.

0:20.1

To learn more about Yachtolt, visit yawcult.co.com.j, that's Y-A-K-U-L-T-C-O-J-P.

0:28.4

When it comes to a guide for your gut, count on YacL.

0:41.4

Happy Monday, listeners.

0:45.0

Let's kick off the week by catching up on some science news.

0:47.3

For science quickly, I'm Rachel Feltman.

0:56.0

Right over here, guys, right over here, right over here. Can we get turn around?

0:59.0

If you enjoyed photos of the red carpet fashion from the Met Gala last Monday,

1:05.0

chances are good you encountered at least one artificial intelligence-generated fake.

1:10.0

Katie Perry posted screenshots

1:12.3

suggesting that even her own mother got duped by an AI-generated image, appearing to show

1:17.6

the pop singer in a floral gown. It just so happens that on Tuesday, chat GPT developer OpenAI

1:24.1

announced a new tool designed to detect images made using the company's Dolly 3 generator.

1:30.7

OpenAI says internal tests found that the tool was able to identify about 98% of images generated

1:37.1

by Dolly 3. The creators did note, however, that any post-AI changes made to the images,

1:43.4

including shifts in coloring, made the

1:45.5

tool more likely to fail. Sam Gregory, executive director of technology-focused human rights

1:50.9

nonprofit witness, told NPR that good media literacy and common sense make a better defense

1:57.4

against deepfakes than currently available digital tools.

2:07.3

While a fake photo of Katie Perry may not pose a global threat, other images could have more serious implications, particularly during an election year.

...

Please login to see the full transcript.

Disclaimer: The podcast and artwork embedded on this page are from Scientific American, and are the property of its owner and not affiliated with or endorsed by Tapesearch.

Generated transcripts are the property of Scientific American and are distributed freely under the Fair Use doctrine. Transcripts generated by Tapesearch are not guaranteed to be accurate.

Copyright © Tapesearch 2025.