Yubo, a leading social discovery app for Gen Z with over 80 million users worldwide, is proud to announce the rollout of SafeBlur, an AI-powered tool that automatically blocks harmful content in direct messages (DMs). By leveraging AI to develop a safety-by-design solution for content moderation in one-on-one interactions, Yubo is effectively addressing a longstanding challenge in user safety and privacy for social platforms industrywide.
Now available in the United States, Canada, United Kingdom, and Australia, SafeBlur also features built-in user reporting capabilities to enable users to seamlessly escalate potential infractions of Yubo Community Guidelines for review and response by Yubo Safety Specialists, thereby driving support for overall content moderation and policy enforcement efforts.
Key features of SafeBlur include:
- Automatic detection and blurring of potentially harmful content in DMs
- Built-in user-reporting flow to drive accountability and support enforcement of Yubo Community Guidelines
- Opt-out function for users who prefer unfiltered communication
SafeBlur was developed to address the distinct challenges of moderating content in private interactions. With fewer observers than in group livestreams, DMs can be a higher-risk environment for harassment or other forms of inappropriate behavior. SafeBlur aims to close this gap by extending to private conservations the same level of scrutiny and protection that Yubo users encounter in group settings. In-depth polling conducted by Yubo in 2024 found that a majority of users across gender groups and ages would use a content moderation tool for DMs, if given the choice.
“SafeBlur’s application of AI to address the longstanding challenge of safeguarding private conversations is a testament to our commitment to online safety innovation,” said Yubo co-founder and CEO Sacha Laizmi. “SafeBlur was created in response to insights we gathered through rigorous data analysis and user feedback, reflecting our proactive approach to solving key challenges in trust and safety for social platforms. We’re proud to offer a solution that protects users and maintains the privacy they expect.”
SafeBlur stands apart from Yubo’s existing Muted Words feature by automatically blocking harmful content, without added user input. As an opt-out tool, SafeBlur gives users the ability to choose for themselves the level of content moderation they want for their private conversations.
As part of the SafeBlur rollout, Yubo will be conducting ongoing evaluations of the tool’s effectiveness and user satisfaction. The company plans to use this feedback to continually refine and improve the feature, ensuring it remains at the cutting edge of online safety technology. The SafeBlur algorithm is designed to be accurate across different languages, ensuring efficacy across the 140 countries where Yubo is active.
