Publication date: 3 November 2026
Australia’s High Court has begun hearing challenges to the law requiring designated social media platforms to take reasonable steps to prevent people under 16 from holding accounts. The case moves the global debate about youth online safety from policy ambition to constitutional design.
Reddit and a separate advocacy group argue that the restrictions impermissibly burden the implied freedom of political communication. The government defends a law intended to reduce exposure to harmful platform features. The court is not being asked to decide whether online harms exist in the abstract. It must examine whether this particular legal mechanism is compatible with Australia’s constitutional system.
A safety rule can create new risks
The law places responsibility on platforms rather than children or parents. Ten major services are currently treated as age restricted, including Facebook, Instagram, Reddit, TikTok, X and YouTube. The regulator says platforms removed access to about 4.7 million accounts around the time the requirement took effect.
Scale alone does not establish success. Age assurance can exclude legitimate users, encourage migration to less visible services and require the collection of sensitive personal information. Australian rules prohibit platforms from forcing people to use government identification or an accredited government digital identity, but companies still need reliable ways to assess age.
Those systems require clear standards for accuracy, data minimisation, appeals and deletion. A young person wrongly blocked from political information should have a practical remedy. A platform should not retain age evidence longer than necessary or reuse it for advertising.
Proportionality should guide the next phase
The strongest policy is one that connects the identified harm to the least intrusive effective control. Different services present different risks. A video feed designed for continuous engagement is not identical to a discussion forum or a channel used for civic information. Regulation should therefore be capable of distinguishing functions, design features and evidence of harm rather than treating every account in the same way.
The law already requires an independent review within two years of its start. Whatever the High Court decides, that review should publish evidence on youth wellbeing, platform migration, false age decisions, privacy incidents and access to news and support services.
Other governments are watching Australia because it moved first at national scale. The lesson should not be reduced to whether a ban survives. The more important question is whether safety regulation can remain effective, proportionate and accountable as technology and user behaviour change.