Terry Flew, Professor of Digital Communication and Culture, The University of Sydney; Australian Research Council Laureate Fellow; Co-Director, Centre for AI, Trust and Governance
This is the second part of a presentation I gave to the LSE Research Dialogue titled ‘Ban, bandwagon or band aid: critical reflections on proposals to restrict children’s access to social media’. The event was held at the London School of Economics and Political Science on 7 May 2026, and other participants were Sonia Livingstone, Eva Liewens and Nick Couldry.
The first part of the post can be found at https://mediated-trust-arts.sydney.edu.au/articles/social-media-minimum-age-legislation-international-impact/
I originally posted this on the final day of the call for submisisons to the UK Starmer Labour Government’s 10-week Growing up in the online world: a national consultation. A lot changed for Sir Keir Starmer and his government over that ten-week period, and the story of who will lead the Labour government and what it will mean for such policies has a while to play out. Put simply, Starmer’s personal opposition to “an Australian-style social media ban” (actually a restriction on minimum age to have a social media account) may not matter, as he may not be the PM with the final say on the issue.
Evidence from Australia after six months
Evidence from Australia after three months of the Online Safety (Social Media Minimum Age Amendment) Act is that outcomes have been ambiguous. A Compliance Update Report released by the Office of the eSafety Commissioner in March 2026 found that while almost half of surveyed parents had at least one child with their own social media account prior to the restrictions coming into effect, this proportion had decreased to nearly one third following implementation of social media minimum age restrictions. Notably, of the parents who reported their child had an account on each platform prior to 10 December 2025, around 7 in 10 reported that their child still had an account, with only 3 in 10 reporting that their child no longer had an account.
This mixed report card has led some to declare the policy to be a failure, while others have identified positive first steps from the new laws. In Australia, analyses range from those seeing the policy as starting to have an impact to those arguing that warnings about the capacity to circumvent age verification requirements were being borne out. The mixed report card in Australia differs from overseas interpretations, particularly in the UK and Ireland, where advocates have been very quick to declare the Australian policy to be a failure. This is motivated in no small part by the current UK government consultation on the topic, and the ongoing tussle between the Starmer Government and the House of Lords on including mandatory age restrictions on social media in the Online Safety Act.
What is becoming apparent is that age assurance technologies are only as effective as the willingness of companies to invest in them and for governments to enforce the regulations that underpin them. The current legislation requires social media companies to take ‘reasonable measures’ to apply age-based restrictions, and the Office of the eSafety Commissioner has the power to issue fines of up to $49.5 million to social media companies that fail to do so. The Office is careful around activating these laws, as adverse judgments are likely to be contested by the well-resourced global tech companies, with considerable cost to Australian taxpayers should their appeals be successful.
As we come to the period six months in from the legislation being enacted, there are some critical policy tipping points emerging. One is that for such a policy to have a significant impact, there need to be positive network effects. In other words, the existence of the laws themselves provides a basis for behavioural change amongst the target population. Research undertaken by the U.S. National Bureau of Economic Research indicates that Australia is some way off such a tipping point.
There have also been amendments made to the legislation which define this social media platforms as being in scope as having algorithmic recommendation engines, infinite scroll, time-limited content or engagement metrics, all of which can drive addictive behaviour. This shift away from including all social media platforms apart from games, messaging and educational apps. This amendment is partly to strengthen the Commonwealth’s position in a pending High Court challenge from Reddit. But it also opens the possibility of companies incorporating safety-by-design principles to be out of scope or developing teen-specific social media apps.
Killing the chicken to scare the monkey: the Australian example
There is a traditional Chinese phrase “Killing the chicken in order to scare the monkey” (杀鸡儆猴). It refers to punishing one individual to make an example of them, thereby intimidating and warning others. It comes from an old folktale about a street performer who earned a living using a dancing monkey. One day, the monkey refused to dance for the crowd. To terrify and discipline the animal, the performer slaughtered a live chicken directly in front of it. The monkey, realizing what could happen to it, immediately started dancing again.
The worst kept secret in the Australian digital tech world is that the companies responsible for implementing age verification for young people’s social media accounts are engaging with the process in such a way that the term desultory would be generous. After the initial big announcement that 4.7 million social media accounts had been closed in January 2026, there has been reporting time and again of young people in the affected age brackets circumventing the age assurance requirements, in some case simply by testing the systems repeatedly until they are approved.
The Age Verification Providers Association (AVPA) has pointed to research by software testing firm KJR that 9 out of 10 social media platforms in scope of the Australian legislation do not even check the age of a new user. Other writers have quoted anonymous industry sources who say that the global leadership of these platforms are quite prepared to pay $49.5 million AUD fines if the policy can be portrayed as a failure to other governments. For the most part, these companies have not engaged third parties who specialise in age verification, instead relying upon internal processes of debateable effectiveness.
This is frequently taken as evidence that the Australian social media minimum age legislation simply has not worked. People and organisations that always opposed such measures have been saying ‘I told you so’.
But to use the Chinese proverb, what if Australia is the chicken that the tech giants are seeking to sacrifice, and other countries are the monkeys intended to be scared off such policies? To continue the metaphor, the biggest simian in the jungle at present is the United Kingdom.
The 10-week consultation announced in March by the (at least for the present) Starmer Labour Government has put the question of an Australian-style social media ban very much on the policy agenda. Having been in the UK during this time, I would say there are more conversations about the pros and cons of the Australian policies in the UK than in Australia where, as I suggested that the LSE Research Symposium, public policy debate is dominated by the ‘Three P’s’: petrol, prices and populism (AKA Pauline Hanson – another P). This has since been joined by a ‘T’, which is tax, so it is PPPPT!
What is looming is akin to the Australian situation when legislation was passed in late 2024, with majority popular support for such a measure and significant opposition from key academics and stakeholder lobby groups. What is more complicated int eh UK case is that the governing Labour Party appears to be split on the issue. Prime Minister Starmer and relevant Ministers such as Secretary of State for Science, Innovation and Technology Liz Kendall and Secretary of State for Education Bridget Phillipson have all spoken against a social media ban. But the three main contenders for the Labour Party leadership – Andy Burnham, Wes Streeting and Angela Rayner – have all supported such a measure. In the face of growing impatience with the co-host of The Rest is Politics Alastair Campbell has described as a ‘never ending process’ on this issue, resolution is pending.
Should advocacy groups accept age-gating and focus on implementation?
The debate about the negative and positive impacts of social media on young people has led to divisive arguments around age-gating as a solution. Yet age-gating, the placing of access restrictions on certain types of content or platforms based on age, is not new. Age-based restrictions on certain types of content have always been a feature of media classification systems. It is interesting to note that there are renewed discussions about introducing classification rules for social media content, which makes it clear that an age-gated Internet remains an option even for those who ostensibly oppose what gets called ‘Australian-style social media bans’.
Certain types of Internet content, most notably pornography, have long been expected to be restricted to those over the age of 18. All pre-Internet media also applied some form of age-gating to access, whether it be restrictions on violent TV content before certain times, age-based access to cinemas for films, or restrictions on the display of content depicting nudity or sexual activity. There was never a decision to not age-gate the Internet or social media. It was simply an outcome that occurred around the world.
In that sense, we can see the popular demand for some form of access controls on social media as less a manifestation of a media-generated ‘moral panic’ and more a demand to reestablish some form of democratic control over global tech giants. It has been odd that those who are most adamant that something needs to be done about the power of ‘Big Tech’ over people’s lives and the fate of liberal democracies have also been among the most vocal critics of social media minimum age rules. But it is not clear why the politics of other proposed measures, such as ‘safety by design’ or a ‘digital duty of care’ will be easier to implement if they are to have substantive impact over platform conduct.
The process of root-and-branch change to how social media companies operate will take a long time to implement, while public expectations are for policies which have clear short-term impacts, and which the public understand. Proponents of an alternative approach also face the dilemma that their own proposals are as likely to be opposed by the major tech companies as age-based restrictions. Given that U.S.-based tech companies are politically close to President Trump, who views all such measures as discriminatory and an attack on American values, there is not a pain-free way of developing policies which challenge platform power.
Social media platforms have already notionally applied a minimum age of 13 for access to comply with the Children’s Online Privacy Protection Act (COPPA) 1998, which was designed to restrict companies from collecting personal data from those under 13 without parental consent. While this does not prevent children under 13 from accessing social media platforms, the reliance of these platforms on data-driven advertising has meant that they have put a minimum age rule in place as an alternative to navigating complex parental consent provisions.
One virtue of legislating age-based restrictions is legal clarity. The self-regulatory minimum age requirements have largely been ignored by social media companies for 20 years, and Meta’s internal documents revealed at the US trials, and previously referred to by Meta whistleblower Frances Haugen, indicate that this was a deliberate choice. Where there is a legislated minimum age for social media use, regulators, companies, children and parents all know what the rules are. Where companies are left with discretion to apply such guidelines, the evidence is clear that they have been ignored.
Age assurance measures are being applied with greater regularity across the Internet. In both the UK and Australia, regulators have applied measures to enforce the minimum age of 18 for access to restricted content, most notably sexually explicit content and violent material. There are various forms of resistance and non-compliance occurring around such measures, such as the Canadian-based site PornHub blocking the accounts of Australian users, or the surge in VPN use in the UK after the measures were introduced.
But age assurance measures will get better as their development becomes a priority. Importantly, Apple has now been applying age assurance technologies to downloading apps from its platforms. Under-blocking (allowing those under the required age access) and over-blocking (blocking those who have a legal right to access the content or service) are always possible. But the idea that children as young as 8 should be accessing the same social media platforms as those aged 16 or over is very hard to make a case for, particularly given what we know from the internal depositions coming out of the U.S. bellwether multidistrict litigation cases against Meta, Google and others.
Looking back on the Australian debates, there was a road not taken by critics of the proposed social media minimum age restrictions. That would have been to accept such a policy but set the age limit at 14 rather than 16. That would have done two things. It would have aligned the Australian standard with existing U.S. laws that ostensibly already apply to the major platforms. It would also have taken a lot of the sting about the debate around implied rights of political communication, as these may apply to 14–15-year-olds, while also recognising that there is such a thing as childhood, and that policymakers have long rated and classified media content around such criteria. It also would have brought the academic community a lot closer to popular opinion on this subject, which has thus far been largely immune to academic arguments that social media minimum age restrictions are morally wrong or practically unenforceable.
