Deep Learning Used To Create Fake Airbnb Listings

Deep Learning Fake Airbnb Listings

Taking a vacation can sometimes be extremely stressful – finding the right, affordable plane tickets, packing up, getting past the airport security hassles, and most importantly, finding a place to stay. In order to escape from the mind-boggling prices and taxes, we are introduced to a website, Airbnb, which lists apartments around the world for rent. Now, a new website which uses deep learning, suggests fake Airbnb listings which might look convincing enough for people to believe they are real.

How can we trust an Airbnb host and their listing? There are a plethora of photos, a detailed description and many reviews previous guests have left which ensure the apartment is the way it was described. Some apartments are even verified by Airbnb, and there is efficient support which makes sure everything gets processed in order. But, what if some fake Airbnb listings are just as accurate and convincing as real ones? What if we can’t tell the difference? Check out the listing below:

“My place is close to Pinewalk Market, Warschauer Strasse, Marcus, Vita National Museum and the Thames. My place is a wonderful residential and lively off Chicago. Public transportation to 1 mile just 35 short blocks to World famous fashion and subway. Close to restaurants and pubs. 3 Blocks to the southside’s CBD and 60 m Ride for 2 people in a large living with its own bathroom, can sleep up to two persons too. The apartment is located close to it the largest district of Woolwich and G.”
This is part of a listing found on a website called This Airbnb Does Not Exist, and like its name suggests, it’s fake. Instead, the website has a deep learning algorithm which generated the apartment on its own. Whenever you enter the website, it will generate a new listing.  More importantly, there are convincing photos which can make it look like the apartment actually exists. Christopher Schmidt, the mind behind this project, was inspired by another viral project called This Person Does Not Exist, which saw the AI and Neural network combined to create accurate photos of people you may think you’ve seen before, when in fact, you haven’t, because they don’t exist.

What’s even more bizarre is that the deep learning algorithm was trained to generate such content by actual imagery posted on real Airbnb websites. The result is more than convincing, a detailed description, the warmth of the speech and photos which show the apartment just the way it’s described.

“This means that just about anyone with a couple hours to kill could create something just as compelling as I did,” Schmidt wrote on the website about his project. “While there are parts of the experience that are weak, overall, I think that it works: the listings are often dubious, but typically plausible enough that they would survive a quick glance.”

That said, rapid AI development can bring more suspicion and fake things than we’ve been used to. Just recently, a research organization, OpenAI, developed an AI text generator which uses small samples of text to create sets of convincing paragraphs usually found in tabloid news. While all these projects are indeed interesting, they do pose a slight fear and paranoia about having to make the difference between what is real and what is fake.

This article was originally published in ValueWalk.

Photo: Yuko Honda