News reports from early June suggest that countries like the United Kingdom and Canada are likely to announce restrictions on social media usage by teens. Australia’s decision to enforce a blanket ban on social media access for users under-16, making it among the first economically advanced democracies to do so, set off a wave of regulatory contagion of countries either attempting to enact or considering various grades of social media restrictions targeted at teenagers and children.
As per a Tech Policy Press tracker, over 40 countries are considering such restrictions, with nearly a quarter of these being from Asia including India, Indonesia, Malaysia, and Vietnam, to name a few.
The enthusiasm among governments to push age assurance measures stand. In many cases, in contrast to the nature of the underlying consensus (whether scientific, or among stakeholders), it can be described as being both confusing and concerning. Moreover, different kinds of age assurance mechanisms (age verification through IDs, age estimation using biometric information, and age inference by analysing behavioural patterns) face various technical challenges and circumvention possibilities which raise broader questions about their effectiveness, proportionality, levels of acceptance, and adverse effects on civil liberties.
Evidence is Mixed, Conflicting and Inconclusive
Sweeping restrictions or bans are particularly fraught because evidence on whether social media usage is outright harmful or beneficial is mixed at best. Some studies show benefits, some underscore the harms, while others fail to establish the link.People arguing for or against them can each cite multiple studies to support their assertions.
Across many years, studies/advisories by the American Psychological Association (2023), the U.S. Surgeon General (2023) cite both benefits and harms. Others from the Royal Society (2023), researchers at the Oxford Internet Institute (2023) did not establish a clear causal link between social media usage, screen time, and well-being.
Recent studies attempting to investigate longitudinal effects across large sample sets in the UK (2025) and Australia (2025) showed a range of effects, including positive outcomes, based on whether usage was light, moderate or heavy. They were also unable to establish a causal link to mental health difficulties.
Meanwhile, localised interventions such as device restrictions in schools/classrooms have, in some cases, been found to improve learning outcomes. But, even here, there is a debate over the extent of benefits, tradeoffs, and the lasting effects of any perceived gains. Such conflicting findings, even for narrower restrictions, reemphasise the need for caution with wide-ranging restrictions that are likely to have many unintended consequences.
Stakeholders Are Divided
Governments are keen to push these through, as just a handful of examples show.
In Malaysia and Vietnam, human rights groups and the press have expressed reservations about consultation processes particularly with respect to age verification and in general, respectively. In India, the landscape is more fragmented. The union government is considering age-based restrictions that could be ‘graded’, marking a shift from its stated position around the time the Australian ban went into effect, after the 2026 Economic Survey of India highlighted the need to tackle ‘digital addiction’.
Additionally, state governments in Goa, Karnataka, and Andhra Pradesh have expressed an intention to regulate access to social media based on age. Ironically enough, India’s data protection legislation may result in the need for age assurance mechanisms thanks to its ostensible goal of restricting processing of minors’ data, which will require verifying who might be a minor likely by processing their data or collecting identification documents.
In France, President Macron pushed for a ban to be fast-tracked, while also using X to thank various countries for “joining the movement”.
In the UK’s case, the recent reports about a social media ban have emerged just weeks after a nationwide consultation on social media restrictions concluded in late May. While, as of this writing, neither detailed responses, nor substantive summaries of responses have been published, data released by the UK government suggests that they may have received over 100,000 responses. Recent reports suggest Prime Minister Keir Starmer reversed an earlier position after facing pressure from parents.
Outside of governments the picture is significantly more complex. There are intersecting debates about the respective roles of the state and parents; amplification of existing social problems and new ones that are emerging as a consequence of direct, instant communication at scale, and whether these are limited to adolescents or should be considered more broadly; whether prolonged usage patterns are a result of habits instead of addiction; and if platform design considerations such as ‘infinite scrolling’, and personalisation (which are common across the internet, including on news portals) can be looked at in isolation from the content itself.
Erik Goldman describes such regimes as “segregate and suppress” and suggests that they open both minors and adults to additional privacy and security risks, and in some cases, even harm minors.
They also impose costs on businesses directly as implementation costs, and indirectly in the form of user drop offs. In March, over 400 security and privacy scientists and researchers signed an open letter urging a moratorium on the deployment of all kinds of age assurance mechanisms until there is scientific consensus on their harms and benefits, and their technical feasibility. Among the many concerns raised, they highlighted the possibility of diminishing safety as children migrate to other platforms, the risks to privacy as lawmakers attempt to restrict tools and methods being used for circumvention, and the possibility of censorship because of the centralisation of power.
In May, 19 signatories, such as Mozilla, Proton, Tor Project, and digital rights groups including Electronic Frontier Foundation, Big Brother Watch, Open Rights Group asked the UK government not to undermine the open web through their “blunt” policy instruments such as bans.
Often missing from such conversations are those subject to these restrictions. Teens in Britain do not seem particularly enthused by sweeping restrictions. They are also actively circumventing them by resorting to techniques such as fake moustaches, VPNs, and using photos of dogs from the internet.
A survey from Australia, as the ban went into effect, showed that they didn’t believe it was enforceable. While they understood some of the risks, they also used social media for learning. A report in India found that over 80 percent had at least one social media account but that a significant portion of their usage was on shared devices, making implementation of any restrictions challenging.
How Effective Has the Ban in Australia Been?
Weeks before it went into effect, a breach that exposed the ID selfies of nearly 70000 Australians confirmed concerns that this exercise would create ‘honeypots’ that hackers would seek to compromise.
In January, just weeks after the ban, the eSafety Commissioner reported that the identified platforms had removed 4.7 million user accounts. In a YouGov survey conducted in January, published in March, nearly 60 percent parents of children under-16 self-reported observing positive behavioural changes such as better relationships and better quality interactions.
However, researchers suggest that the 4.7 million number may be overstated since it included deleted and duplicate accounts. Data reported by Crikey suggested that many teens under-16 retained their accounts.
A report by the Molly Rose Foundation estimated that nearly 60 percent kept access. Even a compliance update by the eSafety Commissioner published in March noted that a “substantial portion” continued to retain their accounts. Academics also pointed out limitations with the YouGov survey, including the presentation of data and its timing during the summer break.
Individuals have reported facing isolation. One survey tracking engagement with the news since 2017 found that approximately half the respondents under-16 reported being affected by the ban and were getting less news as a result.
Meanwhile, an NBER working paper on Australia’s social media ban, which surveyed Australian teenagers between March and April, estimated a compliance rate of 27 percent among 14-15 year old participants, and also cited the absence of individual sanction for minors.
Those who did not comply cited social factors i.e. having peers who continued to remain on social media platforms and a fear of missing out. The authors further estimated that a tipping point for greater adoption among children would be around the 2/3 compliance rate mark. Both the Molly Rose Foundation survey and the NBER working paper report that those circumventing the restrictions found it easy to do so.
The Less-Taken Road We Should Take
Relatively speaking, it is still early days for the Australian ban, but given the regulatory wave sweeping the world, we are likely to have many more data points in the months and years to come.
While the existing evidence points to social factors, political considerations driving such interventions may instead result in politicians seeking more stringent enforcement leading to an escalatory spiral of additional restrictions (VPN restrictions, mandatory device-to-ID linking, etc.) and punitive measures.
On current trajectory, with politics taking precedence over establishing clear evidence, we risk being fixated on the access/technological layer where many deep-seated issues happen to be more visible and perhaps shifting them out of view instead of meaningfully addressing them.
Rather than population-scale experiments, there is a need to invest resources into both deepening our understanding of the nature of problems in different contexts and facilitating a broad-based public conversation with a range of stakeholders.