U.K. Government To Ban Under-16s From Social Media

On June 15, 2026, the Prime Minister announced the government's plans to:

• Ban social media for under-16s. This will cover user-to-user platforms whose purpose is to enable social interaction and which allow users to post material alongside algorithms (messaging services will not be covered).

• Introduce legislation to prevent strangers contacting children on a range of online services including gaming platforms.

• Introduce unspecified measures on livestreams.

• Ban AI 'romantic companion chatbots' and similar intimate functionalities on AI chatbots generally for under-18s.

• Set restrictions to in-scope functionalities to be on by default for 16-17 year olds to prevent a cliff edge when users reach 17.

• Consider overnight curfews and breaks in infinite scrolling for under-18s with more detail on that to be announced in July 2026.

When will the ban take effect?

The government plans to move "at pace" thanks to powers in the Crime and Policing Bill that enable changes to online safety law under secondary legislation. The plan is for legislation to be introduced before Christmas with protections expected to apply from Spring 2027.

How will this work?

The government has not yet set out details about implementation. These are likely to follow in July 2026. It does, however, seem to be relying on age verification. It acknowledges solutions may be imperfect but says this should not be a reason for failing to act.

Questions around whether this is a proportionate and effective solution are, however, unlikely to go away. Many parents support a ban, but it is not just large tech companies that oppose it. It is particularly striking that many charities focusing on the protection of children, including the Molly Rose Foundation (led by the bereaved parents of Molly Rose), have been vocal against an Australia-style ban, given that data shows 60% of Australians under-16 are still accessing social media. The NSPCC also previously proposed actions short of a total ban saying they represented "a practical alternative that would deliver real, long-term protection, not just a temporary fix".

Not only are children (who are often far more tech-savvy than their parents and teachers) able to circumvent age-gating, not least by using VPNs, there are also concerns over the amount of personal data that will need to be processed in order to put age verification in place. It's worth remembering that the government has also called for age verification to be introduced at device level and has said it will legislate on this if its recommendations are not adopted by September 2026. It's possible that this will also be rolled into the planned secondary legislation.

What does this mean for you?

The announcement represents a significant departure from government policy when it launched its consultation on children's safety online in March 2026. At that time, the government favoured letting the Online Safety Act 2023 (OSA) bed in and appeared reluctant to introduce an Australia-style ban. Rumours that any ban would be targeted at a handful of named companies, or solely at specific functionality, proved unfounded and the government itself refers to its plans as "Australia plus".

From a regulatory perspective, this is a bold move which may be open to challenge, particularly as it will be enacted under secondary legislation. While primary legislation is effectively immune from challenge, secondary legislation is subject to review by the courts as it lacks the rigorous multi-stage scrutiny required for statutes (indeed, statutory instruments generally receive very little Parliamentary time). These 'Henry VIII powers' are, in themselves, controversial as they effectively allow the executive to change the law with reduced Parliamentary scrutiny.

The government's consultation only closed on May 28, and while the government was able to publish a preliminary analysis of parental support (among those who responded to the consultation) for a social media minimum age of 16 on June 1, it will not have had time to fully consider all responses across the various questions asked in the consultation in sufficient detail to form a coherent proposal for a ban. The government risks being accused of pressing ahead with a policy it had already pre-determined prior to factoring in all responses to its consultation—something widely considered as a big 'no' for regulatory good practice. Failing to conscientiously consider responses to a consultation can also provide grounds for judicial review and the government could well be challenged on its proposals as a result.

Policy specialists, including UNICEF, claim that safe access to social media brings teenagers important benefits including access to information, freedom of expression and participation, and a vital lifeline for isolated or marginalised youth (including those isolated geographically, disabled children, and LGBTQ+ youth). The type of all-out ban apparently proposed could, therefore, raise human rights issues under the European Convention of Human Rights, including as to whether the eventual measures are proportionate to the aims of the legislation. If secondary legislation were challenged on such grounds, courts have the powers to strike it down, whereas they would only be able to issue a non-binding declaration of incompatibility if the policy is introduced by primary legislation (subject to a further challenge in the European Court of Human Rights).

The new proposals also risk muddying the waters on online protection of children. The OSA already places a duty of care on tech companies to take proportionate steps to protect children from content that is harmful to them. While enforcement actions have been launched, it is too early to understand how the OSA has or will change life online for children. Changing the direction of travel now, without waiting to see whether the existing legislation is working, could create confusion and be damaging as regulators and public bodies will have to change recently formed policies and create new ones. It may also create a vacuum during which tech companies will not know what safety measures to invest in, pending more detail on the new ban. As for possible incremental strengthening of the OSA, in Australia they are now considering introducing a general 'digital duty of care' on social media companies for preventing foreseeable user harms, an approach that does not appear to feature in the U.K. government's thinking.

Overall, the legal basis used to impose a ban, and the lack of strong evidence supporting the argument that a ban is more effective than existing or other measures short of a ban, will make it harder for the government to show that its proposals are reasonable and proportionate and that the benefit of the proposals will outweigh the harm done to those impacted by them. While the ability to use secondary legislation enables the government to move quickly, it also makes its proposals more open to legal challenges. Meanwhile, impacted businesses face another set of complex requirements and further uncertainty.