Tinder tries to curb offensive messages
Tinder parent Match Group is debuting a new screen asking customers "Are you sure?" before they post potentially offensive language. It says the technique has reduced such language by 10% in early testing.
Why it matters: It's another example of the kind of tools tech companies can deploy on the middle ground between just banning content and taking an anything-goes stance.
How it works: When Tinder's AI systems detect language that could be seen as disrespectful, a warning screen pops up, giving users the option to change their wording. The detection algorithm was trained on language that users have reported as offensive in the past.
Between the lines: Tinder says both features appear to be helping, and not just in the immediate moment.
- Those who saw the "Are you sure?" prompt were less likely to be reported for inappropriate messages over the next month.
- Those asked whether they were bothered by content were 46% more likely to report inappropriate messages.
The big picture: Tinder says it's the first dating service to use this approach, though the technique does resemble one used by Twitter.
What they're saying: "As a new generation starts dating, we believe we have a responsibility to educate Tinder's members — many of whom are entering the dating pool for the first time — about what kind of behavior is and is not appropriate when building new relationships," Tinder safety head Tracey Breeden told Axios.
Meanwhile: TikTok is announcing an effort today to allow creators who deal with hate or harassment to more easily deal with large volumes of unwanted comments.