On 3 and 4 November 2026, seven High Court justices heard consolidated challenges to Australia's social media minimum age law from Reddit and the Digital Freedom Project. A ruling is expected to take months. The case will decide more than one statute. It tests whether a government can switch off a whole category of online speech for a whole age group, and treat the result as proportionate.
The strongest case for the ban
The government's argument deserves a fair hearing. According to AP's report of the hearing, government lawyer Joanna Davidson outlined seven risks tied to platform design: addictive use, adverse effects on sleep, depression, anxiety, suicidal ideation, exacerbation of disordered eating, and intensified ADHD symptoms. Her colleague Stephen Lloyd argued the burden on political communication is "indirect in nature and modest in extent." If children really are being harmed by engagement-maximising design, and if platforms have had years to fix it, a legislature acting is not obviously overreach. Parents struggling to police individual apps have a legitimate grievance.
What the law does
The Online Safety Amendment (Social Media Minimum Age) Act 2024 amends the Online Safety Act 2021. As AP reported at the time, it took effect on 10 December 2025 and requires listed platforms to take reasonable steps to keep under-16s off their services. Facebook, Instagram, Kick, Reddit, Snapchat, Threads, TikTok, X, YouTube and Twitch face fines of up to A$49.5 million for failing to do so. The High Court's case page for Reddit, Inc. v Commonwealth frames the question as whether the relevant provisions are invalid, wholly or in part, for burdening the implied freedom of political communication.
Where the law is vulnerable
The constitutional test asks, in essence, whether a measure that burdens political communication is suitable, necessary and adequate in its balance. The pro-innovation objection lands on necessity. The harms the government describes attach to specific design features: infinite feeds, recommendation loops, notifications. A ban removes everything, including features that carry no such risk.
Reddit's counsel, Perry Herzfeld, called the law "ill-targeted" and "over broad," and argued that Reddit should not be on the list of ten restricted platforms at all. That argument matters beyond one company. Reddit is largely organised around text-based communities, and the challengers argue that a ban on accounts also cuts off teenagers' access to civic discussion. The Digital Freedom Project brought its case on behalf of two 15-year-olds, according to the same AP coverage. Their standing as the directly affected group gives the necessity argument a human face that a corporate plaintiff lacks.
The government's "modest and indirect" framing also sits awkwardly with what the law does. An account-level ban on a category of services is not an incidental burden. For a 15-year-old, the practical effect is exclusion from some of the main places where public debate happens.
The evidence problem
Proportionality turns on evidence, and the early record is mixed. AP reported that government data showed 5 million Australian social media accounts had been closed since the ban began, in a country of roughly 28 million people. That is a big number, but it measures accounts, not children protected or harm reduced. The same report says the government acknowledged the ban has not been as effective as hoped, that many young children have slipped through age verification, and that in June the government said it was considering strengthening the laws.
That last point cuts against the government in court. If the measure is both over-inclusive for the speech it blocks and under-effective for the harm it targets, its suitability and balance are both open to question. A law that must be tightened to work, and that is already heavy-handed in its present form, is a weak candidate for the least restrictive means.
There is also a design problem that no judgment can fix. Reliable age checks without collecting identity documents are hard. AP notes that platforms may not demand government ID because of privacy concerns. Pushing providers toward ever more intrusive verification would trade one risk for another, building databases of minors' and adults' identities to protect them from feeds.
A narrower alternative
Nothing here requires accepting that children should be left alone with the worst of social media. The proportionate path is to regulate the mechanisms the government itself identifies. That means default-off autoplay and algorithmic recommendation for minors, restrictions on night-time notifications, stronger duty-of-care obligations enforced by the regulator, and meaningful parental controls. These tools target addictive design directly and leave teenagers' access to news, civic forums and communities intact. They would also be easier to evaluate: regulators could measure changes in time spent, sleep and reported harm, rather than counting closed accounts.
Policymakers elsewhere are watching. The same AP page lists related stories on an EU plan to crack down on addictive social media and AI bot features, and on social media companies suing California over a children's law. The common question is whether to regulate the product or exclude the user. Australia chose exclusion first.
What to watch
The seven justices must weigh the harm of limiting children's political engagement against the harm of social media, as AP summarised it. Three signals will matter. First, whether the court treats a platform-level ban on accounts as a direct or an incidental burden. Second, how much weight it gives to the government's evidence of effectiveness. Third, whether it accepts that less restrictive design-based measures were available. A loss for the government would not end child-safety regulation in Australia. It would force Parliament to write rules aimed at specific harms, which is where good policy should have started.