Meet Kolkata-born Naba Banerjee, Airbnb’s anti-party AI mastermind
Naba Banerjee, the person responsible for Airbnb’s global crackdown on parties, has been on a mission for over three years to combat party “collusion,” flag “repeat party houses,” and develop an anti-party AI system capable of preventing high-risk reservations. “Banerjee is a proud party pooper,” reports CNBC.
Airbnb defines a party as an event that causes significant disruption to neighbours and the surrounding community, taking into account factors like open-invite gatherings, excessive noise, trash, visitor volumes, parking issues, and more.
We’re now on WhatsApp. Click to join
Banerjee joined Airbnb’s trust and safety team in May 2020 and has since implemented various measures to address the party problem, including a ban on high-risk reservations by users under 25 and the rollout of anti-party AI in Australia. These efforts have resulted in a 55% reduction in party reports between August 2020 and August 2022, and more than 320,000 guests have been blocked or redirected from booking Airbnb properties since the global launch of Banerjee’s system in May, the report adds.
With over two decades of industry experience, Naba Banerjee brings a wealth of expertise to her role at Airbnb. Prior to joining Airbnb, she dedicated 13 years to Walmart.com, where her career in product management encompassed various initiatives. These included streamlining package delivery to customers, developing mobile apps to facilitate self-scanning, and expediting the checkout process for shoppers.
Born in Kolkata, India, Naba is a trailblazer in her family as the first female engineer. She currently lives in San Jose, California.
ALSO READ: Bathroom with a bed! California Professor shares Airbnb ordeal, netizens call it original ‘RestRoom’
Banerjee’s personal experiences as a mother of teenagers have helped inform her work in identifying party-related signals, the report adds. Her team has developed sophisticated AI models that analyze hundreds of factors, such as reservation details, user demographics, and location proximity, to assess party risk. The system assigns each reservation a party risk score, leading to either approval or rejection, with additional safeguards for holiday weekends.
While the algorithms have proven effective, Banerjee acknowledges that the world of trust and safety is always evolving, requiring ongoing monitoring and adaptations to stay ahead of parties that aim to circumvent the safeguards.
