Article Lead Image

Super-intelligent Tinder bot uses facial recognition for better matches

Using artificial intelligence to find genuine attraction.

 

Dylan Love

Tech

Posted on Feb 10, 2015   Updated on May 29, 2021, 2:02 pm CDT

There are plenty of Tinder bots, and all of them work the same way: They auto-swipe everything, playing the numbers game. But a new Tinder bot has a smarter idea. 

“While my friends were getting sucked into ‘swiping’ all day on their phones with Tinder, I eventually got fed up and designed a piece of software that automates everything on Tinder,” writes software developer Justin Long.

Long’s software creation, Tinderbox, makes use of facial recognition and lightweight artificial intelligence to make Tinder as hands-off as possible. The craziest part might be that it works really well for him. “I’ve gone on at least 10 dates with the help of the bot and I’ve shown my partners the bot in its entirety. One date literally didn’t believe me and thought I was pulling her leg.”

Justin Long

A user configures the software by answering 60 yes-or-no questions in order to get a sense of what he or she finds attractive in potential Tinder matches. After that, the software goes to work, “swiping right” on faces that it deems mathematically closest to your tastes. If it’s a match, the software will even engage in conversation up to three times before kicking the results to you for manual interaction. You have to be yourself at some point, after all.

Tinderbox's workflow for chatting with your matches.

Tinderbox’s workflow for chatting with your matches.

Justin Long

So how well does it actually work? It’s no big deal to create a bot to interact with an app, but if the results are quality, Long’s tapped into something. “Admittedly, it worked too well and started to conflict with work. Although in a couple cases I had follow-ups and I’m still seeing one person.”

You can read Long’s own writeup of his experiment here.

Illustration by Max Fleishman

Share this article
*First Published: Feb 10, 2015, 2:08 pm CST