Meta adds “world chat” to Horizon Worlds, offering a text-based way of communicating with anyone in the same world session, and promises strict moderation
Meta will automatically scan and remove messages that go against its code of conduct. — After introducing mini-games …
Context & Ripple Effects
World chat is the third leg of Meta's harassment playbook in Horizon Worlds. It started with the on-by-default personal boundary system keeping avatars physically apart in early 2022, followed by the Voice Mode setting that let users garble strangers' voices that June. Text chat closes the loop on the last unmediated channel between people sharing a world session.
Why it matters now: the platform is no longer the same audience those 2022 tools were built for. Since Horizon Worlds opened to 10-to-12-year-olds with parental approval in 2024, and creators began generating worlds through Meta Horizon Studio in 2025, the surface area for unsolicited contact has grown — which makes an always-on, automatically scanned text layer both a feature and a liability.
First-order effects
- Users in any world session can now message each other by text regardless of whether they've interacted before, replacing proximity-only communication with a persistent channel.
- Meta takes on real-time enforcement obligations: every message must be scanned against its code of conduct, shifting moderation from user-triggered tools like personal boundaries to automatic removal at scale.
Second-order effects
- The 10-12 cohort changes the stakes: text chat gives parents a concrete, loggable behavior to supervise where garbled voices and avatar boundaries were diffuse, so Meta's age-rating and supervision systems now have to account for written messages.
- Creators building worlds through Horizon Studio inherit a fixed moderation baseline they can't opt out of, which shapes what kinds of social experiences are viable on the platform.
Third-order effects
- Meta is converging on a governance pattern where default-on, automatically enforced contact rules are the price of admission for opening a social platform to younger and larger audiences — each expansion of access triggers a matching expansion of automated control.
- If this holds, social VR moderation becomes infrastructure rather than a setting: platforms that admit minors will need scan-and-remove pipelines as a core system, not an optional layer, and regulators focused on child safety will treat them as such.
The trend: Social platforms like Horizon Worlds are trading open peer-to-peer channels for default-on automated moderation as their user base skews younger, making trust-and-safety tooling the core product rather than a compliance add-on.