New Zealand's National-led government introduced the Online Safety (Minimum Age and Child Safety Risk Assessment) Bill to Parliament on 24 August 2026, proposing to bar anyone under 16 from holding an account on platforms it labels "high-risk" — Instagram, TikTok, Snapchat and Facebook are named explicitly, with messaging apps, Roblox and AI chatbots carved out. Operators would have to show they took "reasonable steps" to verify age using one of four methods: existing account data, facial age estimation, digital ID services, or formal identification. A self-declared birthdate would no longer be enough.
Education Minister Erica Stanford, who is sponsoring the bill, says a new regulator inside the Department of Internal Affairs would enforce it, with fines up to 10% of a platform's global annual revenue for non-compliance — for a company the size of Meta, a theoretical exposure in the tens of billions. No penalty attaches to children or their parents. A second, quieter provision would require covered platforms to publish an annual, written child safety risk assessment covering every user under 18 — a transparency obligation closer to the EU's Digital Services Act than to a blunt age gate, and one that would matter even if the ban itself is watered down in select committee.
The case for it
The steelman here is real. Regulators across multiple democracies have converged on the same worry: recommendation-driven feeds optimized for engagement produce measurable harms for adolescents — disrupted sleep, social comparison, exposure to self-harm content — and platforms have had a decade to self-regulate without meaningfully changing the incentive. A statutory duty to assess and report on risks to minors, rather than trusting platforms to mark their own homework, is a defensible response even for those skeptical of the age-verification piece. New Zealand is not inventing this argument; it is importing one Australia's Parliament already accepted when it passed the Online Safety Amendment (Social Media Minimum Age) Act 2024, assented to on 10 December 2024.
Why the coalition won't sign on
That Australian law is exactly what is sinking New Zealand's version politically. Both of Luxon's coalition partners have refused support, which is why this is government legislation without a governing majority behind it. NZ First's Winston Peters said his party would "never support this," citing enforcement concerns about VPNs and mandatory digital ID, and calling Australia's ban a "colossal failure." ACT's David Seymour argued the mechanism has no comfortable middle setting: either it is strict enough that adults get swept up in it — he pointed to a French age-verification law that was struck down — or it is "so weak it makes little difference, as Australia has found." Labour has not committed either way, leaving the bill's passage genuinely uncertain despite being introduced by the government.
What Australia's first year actually shows
The opposition's citation of Australia is worth checking against the record rather than taking on faith. Australia's ban took effect 10 December 2025, covering ten platforms including Facebook, Instagram, Snapchat, TikTok, YouTube, X, Reddit and Twitch, with civil penalties of up to A$49.5 million for failing to take reasonable steps to prevent under-16 accounts. Canberra's own account of the results is genuinely mixed. The government says more than 5 million under-16 accounts have been removed, deactivated or restricted since the ban began — a number platforms can point to as evidence of good-faith compliance. But the Communications Minister has separately said she is "not satisfied that tech companies are doing everything they can to keep under-16s off their platforms," and the government has proposed doubling the maximum penalty to A$99 million for systematic breaches, with eSafety actively investigating several of the largest platforms for possible non-compliance.
That is not a law that has settled anything. It is a law nine months into enforcement, still recalibrating its own penalty structure, whose regulator cannot yet say with confidence that the removal numbers reflect durable compliance rather than accounts that migrate, get recreated, or move to platforms outside the ten it covers. Citing it as proof of either success or failure is premature in both directions — which is precisely the problem with New Zealand legislating an near-identical mechanism now rather than waiting a full enforcement cycle to see what Australia's data actually shows.
The proportionality question
The deeper issue for a pro-innovation, evidence-based approach is not whether platforms should owe a duty of care to minors — they should, and the risk-assessment half of this bill delivers that without requiring anyone to submit a passport scan to open an Instagram account. It is that facial age estimation and digital ID checks build verification infrastructure that persists long after the policy debate about teen mental health has moved on, imposes real compliance cost that smaller platforms and new entrants absorb worse than Meta or ByteDance, and — on Australia's own admission nine months in — may not even reliably achieve the stated goal. A regulator that cannot yet say whether its flagship mechanism works is not the precedent on which to build a second country's statute. New Zealand's select committee process, and the outcome of the ACT/NZ First standoff, will be worth watching as a test of whether legislators can separate the parts of this model that are genuinely evidence-based from the parts that are simply popular.