Bumble
has announced Deception Detector, an AI-powered feature that can help identify spam, scam and fake profiles on the platform. The feature is claimed to take action before members even see ‘deception’.
Deception Detector
has been rolled out across Bumble,
Badoo
and Bumble For Friends, as a part of Safer Internet Day.
As per the company, Bumble saw member reports of spam, scam and fake accounts reduced by 45% within the first two months of the technology’s introduction. Citing research, Bumble said that respondents cited fake profiles and the risk of scams as their top concerns when online dating. In India, 29% of respondents said keeping their personal data safer and the risk of scams (28%) as their top concerns when meeting or dating people online.
“Deception Detector is our latest innovation as part of our ongoing commitment to our community to help ensure that connections made on our apps are genuine,” said Lidiane Jones, CEO at Bumble.
“With a dedicated focus on women's experience online, we recognise that in the AI era, trust is more paramount than ever. We are being thoughtful about how to best use new models to reduce the anxiety of making connections and support our community, with AI standing as a main area of focus,” she added.
How Deception Detector works
Deception Detector brings a machine learning-based model to assess the authenticity of profiles and connections on Bumble’s platforms. It said that the feature supported in blocking 95% of the accounts that are identified as spam/scam. This automated and AI-based technology is used in conjunction with dedicated human support to prioritise a safe and empowering community.
Other AI features on Bumble
The company previously introduced other detection capabilities such as Private Detector that utilises AI to detect and blur lewd nude images. Bumble also introduced AI for photo verification and Best Bees, an AI algorithm that “provides a higher level of curation for members”
Bumble For Friends also has AI-powered conversation starters to provide members with a tailored message when reaching out to a new connection.
English (US) ·