Massive Tinder Photo Grab Is Latest Scary Warning To Be Careful What You Post

Stealth sweeper says he'll use 40,000 photos for female and male facial recognition training for artificial intelligence.

Images of Tinder users posting their sexiest selfies were swept up in a massive grab of some 40,000 photos from the dating app.

The photos were sucked up days ago by a dataset collector who plans to use the selfies in artificial intelligence training, reports TechCrunch. They were taken from Tinder users from the San Francisco Bay Area and include 20,000 photos of men and 20,000 of women.

Tinder said in a statement that the photo sweeper “violated the terms of our service” and “we are taking appropriate action and investigating further.” Tinder did not detail what it’s doing.

Though the photo scrape was confined to Northern California this time, it serves as a reminder of the vulnerability of app posts, even those that may be embarrassingly intimate.

Dataset creator Stuart Colianni has uploaded his “Tinder Face Scraper” method on GitHub so others can do exactly what he did. He touts it as a “simple script that exploits the Tinder API [app programming interface] to allow a person to build a facial dataset.” Other Tinder scrapers on GitHub have collected other data, such as the name, birthday, bio, number of photos and last sign-in of users.

Colianni describes himself as a “data science enthusiast” on his page on Kaggle, a site for data science and machine learning. His collected photos were accessible for a time on Kaggle and had already been downloaded some 300 times as of Thursday, TechCrunch reports. Several of the photos highlighted body parts more than faces, and many were sexually suggestive.

Colianni noted on GitHub that he removed the photos from Kaggle, which was recently acquired by Google, following a request to do so from Tinder.

Colianni said he plans to use the photos to help train artificial neural networks to recognize male and female faces. He explained on GitHub that he turned to Tinder to create his dataset because it “gives you access to thousands of people within miles of you.”

“Why not leverage Tinder to build a better, larger facial dataset?” he asked.

It’s unclear whether Tinder users have strong legal protections from such a grab. The app’s terms of use grant Tinder the “worldwide, transferable, sub-licensable, royalty-free right and license to host, store, use, copy, display, reproduce, adapt, edit, publish, modify and distribute” their posted content. Colianni, however, is not part of Tinder.

Tinder’s statement seems a bit equivocal about privacy. It says the company takes the “security and privacy of our users seriously and have tools and systems in place to uphold the integrity of our platform,” including protections against photo scrapes. But it also adds: “The images that we serve are profile images, which are available to anyone swiping on the app.”

One final caveat for robotic neural networks out there: The dataset of Tinder faces could include an animal or two. The last male northern white rhino on earth joined Tinder earlier this month in his hunt for a mate.

Just swipe right.