Airbnb's 'Party Pooper' Has Had Great Success

Airbnb head of trust and safety Naba Banerjee might be a world champion in thwarting parties. Banerjee—described by CNBC as a "proud party pooper"—helped reduce reports of parties at Airbnbs by 55% over two years. A reservation screening system introduced in May has already blocked 320,000 bookings, according to Airbnb. The company announced a "party prevention system" at the start of the summer to further reduce disruptive and unauthorized gatherings. The company defines a party as a gathering at one of its listing that "causes significant disruption to neighbors and the surrounding community."

A ban on parties introduced after a series of high-profile incidents, including a shooting that killed five people in California in 2019, was made permanent last year. Banerjee tells CNBC that as a mother of five, she has a knack for knowing when an unauthorized party is in the works. "Being a mother of teenagers and seeing teenage friends of my kids, your antenna is especially sharp and you have a radar for, 'Oh my God, okay, this is a party about to happen,'" she says. "Between our data scientists and our machine learning engineers and us, we started looking at these signals."

In an artificial-intelligence screening program first introduced in Australia, Banerjee's team trained algorithms to consider hundreds of factors, including the user's age, when determining how much of a party risk a booking was. Banerjee says the system assigns a level of "party risk" to each booking. A booking by a user under 25 with an account created at the last minute would be considered risky, especially if it involved an entire house rented for a single night on a holiday weekend, and if it was close to the user's home address. She says the system was trained on information about problematic booking, as well as examples of "good" Airbnb customers. (More Airbnb stories.)