Analysis: TikTok's algorithm actively promotes child begging livestreams and TikTok profits from the content, with an up to 70% cut, despite anti-begging rules
www.theguardian.com/technology/ 2...
Context & Ripple Effects
TikTok previously said it would act promptly on exploitative begging after reporting on refugees seeking digital gifts; this investigation tests whether that commitment translated into enforcement on Live. Its recommendation system has also been described as optimizing for retention and time spent in an earlier account of the algorithm's engagement goals.
The report adds a monetization dimension to an existing safety record: a Utah lawsuit had alleged that TikTok Live promoted sexual streams to minors and enabled money laundering. The common issue is whether Live's recommendation and gift systems create incentives that stated rules do not adequately constrain.
First-order effects
- Accounts streaming children begging can receive algorithmic distribution and digital gifts despite TikTok's anti-begging rules, while TikTok retains an up-to-70% share of the reported transactions.
- TikTok faces immediate pressure to explain the gap between its enforcement policy and the treatment of the implicated accounts, 14 of which it says do not violate its rules.
Second-order effects
- The findings put TikTok's Live moderation, recommendation, and gift-revenue controls under closer scrutiny, especially following allegations concerning harmful Live streams shown to minors.
- If enforcement removes these streams, affected broadcasters lose a distribution and income channel; if it does not, the platform's safety assurances become harder to separate from its transaction incentives.
Third-order effects
- The case points to a broader platform-governance problem: when recommender systems and virtual-gift revenue reward the same content, content rules need auditable enforcement rather than policy statements alone.
- Sustained evidence of this mismatch could make Live recommendation and payment design a more central focus of child-safety oversight, alongside moderation of individual streams.
The trend: This is one data point in the shift from judging social platforms by published rules toward judging them by how recommendation and monetization systems operationalize those rules.