At Facebook, the fight against fraud and disinformation is done more and more thanks to AI (artificial intelligence). According to American media, the social network has developed a new machine learning model capable of detecting false accounts. Last year, this software would have removed more than 6.6 billion false identities on the platform. It’s enormous.
Facebook has a new AI tool to spot spammers, and it’s already taken down billions of accounts
To achieve this, this artificial intelligence examines more than 20,000 “deep features” of each account. It is not only raw data entered by the user, but above all behavioral data. With whom does it communicate? How frequently? How many invitations does he send? In which groups is he active? how does he behave in these groups? etc.
Gizchina News of the week
The complexity of this model allows Facebook to outstrip the vast majority of fraudsters, in any case for now. Facebook currently estimates that about 5% of its 2.98 billion active users are fake people. And every day, it continues to block millions of attempts to create fake accounts. An endless fight.
It is worth mentioning that recently, Facebook has announced the cancellation of F8, its annual developer conference. This was to take place on May 5 and 6, 2020 in San Jose.
“F8 is an incredibly important event for Facebook (…) but we must give priority to the health and safety of our developer partners, our employees and all those who help in the realization of F8”, explains Konstantinos Papamiltiadis, director of partnerships at Facebook, in a blog note.
Each year, the F8 conference is an opportunity for Facebook to present new products, technologies and, of course, development tools. This cancellation is therefore a loss of valuable information. This is why Facebook is looking for alternatives. The company plans to organize events at the local level in addition to Livestream events.