A read-only account, not a lockout
Vietnam's Ministry of Culture, Sports and Tourism circulated a draft decree, presented at a Hanoi public-consultation workshop on July 23, 2026, that would bar social media users under 16 from posting, commenting, sharing or reacting to content — while still letting them view it (VnEconomy). The draft amends Decree 147/2024/NĐ-CP, the internet-services law the government issued on November 9, 2024. Under-16 accounts could not be self-registered; a parent or legal guardian would have to register and supervise them. Platforms — domestic and foreign — would need "technical measures" to identify child users and route them to age-appropriate content. A parallel provision caps online gaming for under-16 users at 60 minutes a day per publisher (Reuters via The Star). Deputy Culture Minister Phan Tam framed the goal as protective rather than exclusionary: "The primary objective is not to ban or excessively restrict children's access to social media, but rather to ensure that when children use social media services, they are placed in an age-appropriate environment and protected from inappropriate risks." The text is not final and can still change before adoption.
The draft sits inside a broader push. In March 2026, Prime Minister Pham Minh Chinh signed Decision 468/QĐ-TTg, a 2026–2030 national program for children's online protection built around universal school cybersecurity tools, ISP-level content filtering and digital-literacy campaigns (Chinh Phu / Bao Chinh Phu). The under-16 posting restriction is the sharpest tool yet produced under that mandate.
The case for it
The strongest version of the government's argument deserves a fair hearing. Vietnam had roughly 79 million active social media identities as of October 2025 — 77.6% of the population — on a platform mix (Facebook, TikTok, Zalo, YouTube) that skews young and algorithmically driven (DataReportal). Cyberbullying, unsolicited contact from adults, and compulsive short-form video use are documented harms, not moral panic. And a read-only model is genuinely more proportionate than the alternatives on the table elsewhere: a child who cannot post is still not cut off from information, schoolwork coordination, or family contact, which is more than can be said for a hard lockout. Requiring guardians to register accounts, rather than banning them outright, keeps the family — not the state or the platform — as the primary supervisory node. That is a defensible design choice, and Hanoi deserves credit for not simply copying the blunter models now spreading through Europe and the Asia-Pacific.
Why the blunter models are the wrong benchmark
That comparison is also the draft's biggest liability, because the blunter models are not working. Australia's under-16 ban — the world's first, in force since December 10, 2025, with fines up to A$49.5 million for non-compliant platforms — has already needed a fix. A University of Newcastle study found that more than 85% of Australian children under 16 were still using social media three months after the ban took effect, prompting Canberra in June 2026 to double the maximum penalty to A$99 million and give the eSafety Commissioner power to compel internal company documents (Xinhua). Prime Minister Anthony Albanese's own diagnosis — that platforms are not "doing enough" — describes a compliance gap, not a policy solved by statute. France has since passed a similar under-15 restriction, and a lengthening list of countries — Austria, Denmark, Canada, Indonesia, Malaysia among them — are drafting comparable rules (TechCrunch). Vietnam's decree is being written into a policy environment where the flagship precedent is already straining.
Vietnam's version is more surgical than Australia's, but it does not escape the same enforcement mechanics. A parent registering a child's account can still hand over the login. "Technical measures to identify child users" is a mandate, not a solved engineering problem — age-inference and liveness-check systems remain porous to VPNs, borrowed IDs and cross-border app stores, the same gaps Australian regulators are now trying to close after the fact. Writing the identification burden into law before the technology is proven risks locking in either weak, easily-evaded checks or invasive ones — biometric or ID-linked verification that creates its own data-protection exposure for exactly the population the decree is meant to protect. Decree 147 already requires phone-number or national-ID-linked authentication for anyone posting on Vietnamese social media; layering child-specific biometric or ID checks on top raises the stakes on a data store that becomes a more attractive target the more sensitive it gets.
What a proportionate final text would do
The decree is still a draft, and that is exactly when the design choices are cheapest to fix. Vietnamese regulators should build in what Australia's rushed rollout skipped: a defined, narrow set of acceptable age-verification methods that does not default to biometric collection or permanent ID linkage; a published compliance-rate metric, reviewed on a fixed schedule, so the rule can be judged by outcomes rather than intent; and proportionality between platform size and compliance burden, so the identification mandate does not simply become a cost that only Meta- and ByteDance-scale operators can absorb, freezing out smaller and domestic competitors. A read-only model for minors is a more thoughtful starting point than an outright ban. Whether it protects children or just adds friction depends entirely on enforcement details that remain, as of this draft, unwritten.