What parents need to know about Bluesky, the decentralised Twitter-style network gaining UK teen users, including the AT Protocol, the labelling system, and how moderation differs from X/Twitter.
Bluesky is a decentralised social media app similar to X, where users post short public updates and follow feeds, built on an open network.
Official age
13+
We recommend
16+
Developer
Bluesky Social PBC
Risks
4
Bluesky is a public microblogging service built on the AT Protocol, an open standard the company developed so that accounts, posts, and recommendation feeds can move between providers. The app feels similar to early Twitter - short text posts, replies, reposts, and quote posts - but the moderation architecture is unusual. Bluesky enforces a baseline against illegal and harmful content, and on top of that any community group can run a 'labeller' which tags content as, for example, adult, political, or graphic. Users choose which labellers to subscribe to and what each label should do: hide, warn, or show. This gives older users fine-grained control, but it also means a child's experience depends heavily on which feeds and labellers they follow, and there is currently no UK family-pairing tool.
UK teenagers are using Bluesky as a calmer alternative to X/Twitter, especially for fandoms, art and writing communities, queer support spaces, and tech and news commentary. Some follow curated 'custom feeds' for specific interests rather than relying on the default discover feed. Many bring their handle from X across, and a small number cross-post the same content to both. Because the feed can include adult-rated posts from accounts the child has not personally followed - via reposts, quote posts, or recommended feeds - children can encounter unmoderated material faster than on a platform with stronger algorithmic safety rails.
Adult Content
Location: Settings -> Moderation -> Adult Content
Set to: Disabled for under-18s
Bluesky requires users to confirm they are 18+ before enabling adult content. Keep this toggled off on a teen's account.
Content Labels
Location: Settings -> Moderation -> Content Filters
Set to: Hide adult, graphic, and intolerance categories
Set each label category to 'Hide' rather than 'Warn' so the post is filtered out completely instead of behind a tap.
Logged-Out Visibility
Location: Settings -> Privacy and Security -> Logged-out visibility
Set to: Hide profile from logged-out users
Stops scrapers and unauthenticated users from seeing your child's posts in search and on third-party AT Protocol clients.
Direct Messages
Location: Settings -> Privacy and Security -> Direct Messages
Set to: From users I follow
Limits DMs to accounts the child has chosen to follow, blocking unsolicited messages from strangers.
Subscribed Labellers
Location: Settings -> Moderation -> Subscribed Labellers
Set to: Bluesky default only
Avoid adding community labellers you do not understand - some are run by partisan or adult-content groups whose 'warnings' can backfire.
Sit with your child while they configure the Moderation menu and set adult, graphic, and intolerance labels to 'Hide'
Time: 15 minutes
Explain how custom feeds and labellers work, and agree not to add new labellers without discussing them first
Time: 15 minutes
Talk through how quote posts and reposts can pull strangers' content into your child's feed even when they have not followed those accounts
Time: 10 minutes
Agree a no-real-name, no-school, no-precise-location rule for the profile bio
Time: 5 minutes
In immediate danger: call 999. For non-emergency police matters, call 101.
Concerned about a child but it's not an emergency? NSPCC helpline 0808 800 5000. Childline for young people 0800 1111.
Myth: Bluesky is decentralised like Mastodon, so the same 'pick a good server' safety advice applies.
Fact: Unlike Mastodon, almost all Bluesky users are on the company's own hosted service, even though it runs on the open AT Protocol. There isn't yet a real ecosystem of independent Bluesky servers to choose between, so the safety-relevant choice for a child's account today is which labellers and custom feeds they subscribe to, not which server they join.
Myth: Bluesky now runs age checks in the UK, so my teenager's account is fully age-verified and safe.
Fact: UK age checks on Bluesky are used to gate specific things, such as unlocking direct messages or adult-rated content, rather than certifying the whole account as safe for a child. Confirming an age does not change what a teenager can still encounter through reposts, quote posts, or recommended feeds, so the moderation and labeller settings still need checking directly.
Myth: Turning on Bluesky's content labels works like a parental control - if filtering is on, my child won't see anything inappropriate.
Fact: Labels depend on community-run labellers flagging content after the fact, which is inconsistent rather than preventative, so adult or graphic posts can circulate before anyone labels them. A teenager can also subscribe to additional labellers themselves, which can loosen rather than tighten what gets shown.
Bluesky is not built for this age at all - it's an open, adult-oriented public feed well below the app's own minimum age of 13. If a child hears about it from an older sibling or friend, it's a good moment to explain in plain terms why an open network can be harder, not easier, for one company to fully keep an eye on.
Below Bluesky's minimum age, and the open feed design means content can arrive from accounts and labellers a child never chose to follow. If they're curious, point them to a more contained app for now and revisit Bluesky once they're old enough to understand labellers and custom feeds.
Workable only with hands-on setup: sit together to confirm adult content is switched off, content labels are set to hide rather than warn, and no extra labellers have been added without talking it through first. Check back occasionally, since a teenager can quietly add a new feed or labeller that changes what shows up.
Most older teens can manage their own moderation settings, but it's still worth talking through how quote posts and reposts can pull a stranger's content into a feed unasked, and agreeing they'll flag it if a labeller or feed starts surfacing something troubling.
Practical updates on apps and online safety. No spam.