With the Fourth of July weekend comes fireworks, hot dogs, parades and parties.
For Airbnb, the short term rental platform, these parties are not exactly a fun time: They are a scourge of noise complaints, injury risks and property damage.
To counter these disruptions and help enforce the platform’s year-round ban on open-invite parties, Airbnb five years ago began using machine-learning technology to weed out what the company describes as “higher-risk” bookings. The technology will likely be working overtime during the country’s 250th birthday celebration next week.
The technology, the company says, uses “hundreds of signals” — such as whether a user is making the reservation locally or for a short one- or two-day stint — to flag whether a booking is characteristic of the kind of reservations that result in a rental turning into a party house.
Guests who are, as a result, prevented from booking an entire home, are instead redirected to book private room listings and hotels on the Airbnb platform.
Last Fourth of July, the company’s system deterred more than 20,000 people in the U.S. from booking an entire home listing on Airbnb, according to company spokesperson Ruthie Kongo.
Around 2,500 people in California were redirected, and in Los Angeles, specifically, some 200 people were deterred, she said.
“These efforts reflect our ongoing commitment to help reduce the risk of disruptive parties, and we are seeing positive results,” Airbnb said in a statement on its website.
In 2025, fewer than 0.06% of stays on Airbnb in the U.S. resulted in a report of a party, the company said. Company filings say Airbnb recorded 533 million nights and seats booked in 2025, but did not break out property reservation numbers for the U.S.
Param Vir Singh, a professor of business technologies and marketing at Carnegie Mellon University, sees the anti-party system as a useful tool to screen bookings at scale, given that the company “can’t have a person sitting and evaluating every booking to see if it will lead to a party.”
But Singh said that any machine-learning tool using vast amounts of consumer data to make decisions may have blind spots or false positives. A case in point is Singh’s study of racial disparities in the way hosts use the company’s free algorithm-based smart-pricing tool to set their properties’ daily price automatically.
Airbnb executives say they hope to make deeper inroads with such tech and have various ventures in the works.
During an earnings call in May, Chief Executive Brian Chesky said AI now writes nearly 60% of new code, produced by the company’s engineers.
“AI gives huge leverage,” he said on the call.
Chesky is also in the early stages of funding for a new AI lab he plans to create separate from Airbnb, Bloomberg reported. Chesky will remain chief executive of Airbnb and will not take a central role leading the new lab, according to Bloomberg.