Bumble’s Testing Reveals AI Tool Successfully Blocks 95% of Malicious Accounts Automatically
Bumble, the popular dating app, has released a new artificial intelligence (AI)-based feature to enhance the safety of its user base. The AI tool is called Deception Detector and it can identify and block spam, scams, and fake profiles before members come across them. The feature was launched to mark the Safer Internet Day, which falls on February 6. The dating app claimed that during the testing phase of the new technology, user reports of spam, scams, and fake accounts were reduced by 45 percent.
Bumble said in a press release that it conducted a survey last year with a global sample size of 28,000 people and found that fake profiles and scams were the top concerns of the respondents. In India, 29 percent of the surveyed individuals highlighted wanting to keep their personal data safer as the top concern, while 28 percent said the risk of scams when meeting someone for the first time took priority, the company stated. The findings from this survey led to the building of Deception Detector.
The AI tool uses a machine-learning-based model to measure whether a profile and its connections on the dating app are authentic or not based on several parameters. If a profile has been identified as a spam or scam account, the tool automatically blocks it. The dating app claims that 95 percent of accounts that showed up in its testing as fraudulent were already automatically blocked by Deception Detector. However, it is being used alongside dedicated human support to minimize the risk of any malicious accounts escaping its reach, or innocent accounts being banned accidentally.
“Bumble Inc. was founded with the mission to cultivate equitable relationships and empower women to initiate connections, and the Deception Detector represents our latest innovation in fulfilling our ongoing commitment to our community, ensuring that interactions on our platforms are authentic. With a dedicated emphasis on enhancing women’s online experiences, we understand that trust is more crucial than ever in the era of AI,” stated Bumble CEO Lidiane Jones. Jones also emphasized the company’s continued focus on AI development.
The Deception Detector is not the first AI tool introduced by the dating app. In 2019, Bumble launched the Private Detector AI, which automatically blurs explicit photos sent by users and provides an alert. Later, the company released an open-source version of the feature on GitHub for the broader community.