Not every app arrives with a cautionary tale attached as Yubo did. Figuring out what the app is today requires understanding what it was when it launched under a different name, with different rules, and almost no real moderation to speak of.
In 2015, three French developers in their early twenties built an app called Yellow. The stated intention was a social discovery platform for teenagers, a space for young people to meet others their age at a time when Facebook was becoming their parents’ platform and Snapchat had its own set of problems.
The mechanic they chose, however, was the one that Tinder had made popular: swipe right to connect, swipe left to move on.
On a dating app, that works because the intent is clear, but on a platform populated primarily by 13 to 17-year-olds, it created problems the founders either did not anticipate or did not adequately prepare for.
Yellow and What Went Wrong
Yellow launched globally in 2016 and grew rapidly, reaching tens of thousands of downloads a day at its peak and well over 7 million users by mid-2017. The demographic was almost entirely teenagers, with the fastest growth coming from the 14 to 16 age group.
The app asked users to enter a birth date to separate teens from adults but required no ID verification, meaning the date could be set to anything. Adults could enter the teen side of the platform. Teenagers could enter the adult side. Neither group was stopped.
The NSPCC warned parents about Yellow in 2016. By 2017, news articles ran pieces on it, with headlines calling it “Tinder for teens.” Schools began advising parents to check if their children had the app installed.
What was meant to be a friendship platform had acquired the reputation of something else entirely.
The Rebrand and the Live Stream Pivot
In December 2018, Yellow became Yubo. The founders wanted distance from the Tinder-for-teens label and restructured the product around live streaming as the core mechanic, with group connection rather than one-to-one matching as the stated goal.
Up to ten users could co-host a live stream while hundreds more watched and commented. Streams were separated by age group, public by default within those groups, and scrolled rather than swiped.
The intent was to shift the dynamic from intimate pairing toward something closer to a digital hangout.
By 2019, Yubo had 20 million users, and by the end of 2020, that number had doubled, partly driven by pandemic-era social isolation. Yubo has raised over $60 million in total funding, including a $47.5 million Series C round backed by venture capital firms.
Co-founder Sacha described Yubo as a place to “hang out live” rather than scroll through a content feed. The platform had no likes or followers and did not rely on a traditional algorithm.
Critics, however, argued that changing the way the platform worked on the surface did little to address its bigger moderation problems.
The Turning Point
In May 2022, a gunman responsible for the mass shooting at an elementary school in Uvalde, Texas, was an active Yubo user in the weeks before the attack.
He reportedly spent hours on the platform each day and made threats that other users could see and reported several times. His account remained active despite those reports.
The backlash was severe and immediate, reaching members of the US Congress, and Yubo could no longer argue that its moderation was adequate.
What followed was the most significant safety overhaul in the app’s history, including expanded harmful language detection, improved reporting tools, mandatory biometric age estimation introduced in 2023 for all new and existing users, and phone number verification.
Tens of thousands of accounts were removed when the age verification was enforced at scale. The Australian eSafety Commissioner, which had been tracking Yubo critically for years, publicly recognized the app as a platform that had implemented genuine age estimation alongside other considered safety features.
Human Moderation Versus AI Moderation
The Yubo story is a useful lens on a question that every large social platform eventually confronts. Human moderators are better at context; they can read subtext, recognize patterns, and make judgment calls about borderline content.
They are, however, expensive, inconsistent, and impossible to deploy across every corner of a platform serving tens of millions of users. AI moderation scales, but it consistently misses what it was not specifically trained to find.
Yubo’s earlier AI nudity filters, for instance, could not detect suggestive roleplay or indirect threats. Human moderators were deployed only on the most popular live streams, leaving the smaller ones unwatched.
The approach the app has landed on is a layered one, with AI handling detection at scale, human moderators handling escalation and edge cases, and structural safeguards like age verification reducing the population of bad actors before they enter the platform at all.
It is not a perfect system. It is a more honest one than what existed before.
What the Numbers Reflect
On the Play Store, Yubo holds a 4.6 rating from 415,000 reviews and has crossed 10 million downloads. On the App Store, it sits at 4.5 from 373,000 ratings and ranks 32nd in Social Networking.
Worth noting is that the App Store lists an 18+ age rating with in-app controls, which sits in some tension with a platform that still maintains separate age groups for users under 18.
The overall trajectory of those ratings, sustained and positive across a very large user base, suggests the product has genuinely improved from where it started, even as the criticism around facial recognition and residual moderation gaps has not entirely disappeared.
Yubo now has over 60 million users. The journey from Yellow to that figure is a decade-long case study in what happens when a social platform for young people grows faster than its safety infrastructure and what the cost of closing that gap eventually looks like.
Who Is It For?
Yubo is a platform for young adults looking to meet people their age through live streams and group chats rather than curated posts and follower counts.
The absence of likes and algorithmic amplification is a genuine structural difference from most social media, and the live format creates a kind of connection that a feed does not. For parents considering whether a teenager should be on it, the safety improvements are real and documented.
Whether they are sufficient is a judgment call that the platform’s history makes it reasonable to think carefully about.
