UK social media curfews for 16 and 17-year-olds explained

What the government has announced

The Department for Science, Innovation and Technology confirmed on 15 July 2026 that 16 and 17-year-olds in the UK will get default overnight curfews and automatic restrictions on addictive design features on social media. Under the plan, apps will switch off access by default between midnight and 6am for this age group, and features built to keep users scrolling, such as videos that autoplay one after another and feeds that continually refresh with personalised content, will also be disabled by default. Older teenagers will still be able to change these settings themselves if they choose to, so the restrictions function as a stricter default rather than a hard block.

Technology Secretary Liz Kendall framed the package as a response to consultation feedback from parents and teenagers, who said that even as 16 and 17-year-olds gain more independence, they should still be shielded from the platform features most associated with harm to sleep, concentration and family time. Alongside the social media changes, the department also set out an intention to bring forward separate protections for children using AI chatbots, including regular mandatory breaks for under-18s and tighter scrutiny of chatbots that offer mental health advice, up to and including bans on services judged to pose a serious risk to children.

Filling the gap left by the under-16 ban

The 16 and 17-year-old measures are explicitly designed to sit alongside the government's wider ban on social media accounts for under-16s, due to take effect from spring 2027. DSIT's stated concern is that without further protection, a young person would face a sudden loss of safeguards the moment they turn 16 and become old enough to hold an account under the new regime, moving overnight from no access at all to unrestricted access to the platforms' most engagement-optimised features. The curfew and feature restrictions are intended to smooth that transition by keeping some structural protection in place through the later teenage years, while still giving 16 and 17-year-olds meaningfully more autonomy than under-16s.

The evidence behind the curfew model

The policy rests substantially on a DSIT-commissioned qualitative study carried out by Savanta, which ran 309 in-depth interviews with young people aged 13 to 17 and their parents or carers before a one-month intervention period in May 2026, followed by further interviews in June. Participants were split across a control group and three intervention groups, each trialling a different restriction on six core platforms (Snapchat, TikTok, Instagram, Facebook, X and Reddit): a 15-minute daily limit per app, a 9pm-to-7am curfew, or complete removal of the apps.

Each model produced a different balance of benefits and costs. The daily limit encouraged more deliberate, self-rationed use but generated the most friction, since access could end mid-conversation, and it had the lowest compliance of the three. Full app removal produced the largest fall in social media use and the clearest gains in focus and family time, but also caused the most social disruption and the sharpest initial emotional distress, and was widely seen by participants as too extreme to sustain long-term. The overnight curfew sat between the two: it was the easiest to set up and maintain, produced the strongest and most consistent sleep improvements of the three models, and was the version families were most likely to say they would keep voluntarily after the study ended. Its main limitation was that it left daytime and evening use largely untouched, with many participants simply shifting their activity to just before or just after the restricted window.

The researchers were careful to describe the study as qualitative and exploratory rather than statistically representative, and noted that compliance was imperfect across all three groups, with device-switching, parental exceptions and shifting the timing of use all common workarounds. The report's authors concluded that restrictions were most likely to stick where they were framed as a collective, universal norm rather than an individual imposition, paired with education rather than technical controls alone, and developed with young people rather than imposed on them.

Why defaults, not just age checks

A second DSIT-commissioned study, carried out by BMG Research with a nationally representative sample of 2,299 children aged 11 to 17, helps explain why the government has leaned on default settings rather than relying solely on age verification at the point of access. The research found that 78% of children have encountered some form of age check, but the checks they meet most often, ticking a box or entering a date of birth, are also the ones they rate as least effective: 54% of those who had ticked a box thought it did not work. More advanced checks such as government ID upload or facial age estimation were rated far more effective, at 86% and 73% respectively, but were also encountered far less often.

The same research found that 39% of 11 to 17-year-olds have successfully got around an age check at least once, most commonly by giving a false date of birth, and that 26% have used a VPN, rising to 31% among 16 to 17-year-olds. Among children who had used a VPN, 56% had successfully bypassed an age check, compared with 33% of those who had never used one. The study also linked bypassing behaviour to exposure to harmful content: 51% of children who had got around an age check reported seeing at least one type of harmful content afterwards, and explicit sexual content showed the strongest link to bypassing of any content type tested. Because age checks alone can be routed around, particularly by older, more digitally confident teenagers, building restrictions into a platform's default settings removes some of the reliance on a check being passed correctly in the first place.

AI chatbots and the wider media literacy push

The July announcement extends the same protect-by-default logic to AI chatbots. DSIT intends to bring forward measures requiring regular breaks for under-18s using chatbots, and to work with regulators on services that give unverified or potentially dangerous mental health advice, with an outright ban considered for the highest-risk cases. This sits alongside a broader expansion of the government's Kids Online Safety Hub and changes to what children are taught in school. From September 2026, Relationships, Sex and Health Education lessons will include critical thinking about AI and chatbots, spotting mis- and disinformation, and recognising violent or misogynistic content. A wider overhaul of the National Curriculum from September 2028 is intended to embed media literacy across subjects, alongside changes to English, History and computing content covering AI, data science and bias.

Timeline and what happens next

The government has said the first set of regulations covering the 16 and 17-year-old restrictions will be laid before Parliament by the end of 2026, with the measures expected to come into force in spring 2027, timed to align with the separate under-16 social media ban. Implementation and enforcement detail, including how platforms will be expected to identify and treat 16 and 17-year-old accounts differently from adult accounts, has not yet been published. Businesses operating consumer-facing platforms likely to be used by this age group, and parents currently navigating existing parental controls, both have a period ahead in which the practical rules are still being finalised rather than in force.

Frequently asked questions

When do the new social media curfews for 16 and 17-year-olds start?

The government intends to lay the first regulations before Parliament by the end of 2026, with the measures expected to come into force in spring 2027, alongside the separate ban on social media accounts for under-16s.

What exactly will be switched off by default?

Social media access will default to being switched off overnight, between midnight and 6am, for 16 and 17-year-old accounts. Features designed to extend use, such as autoplay between videos and continually refreshed personalised feeds, will also be switched off by default. Older teenagers will be able to change these settings themselves.

Can a 16 or 17-year-old turn the curfew or feature restrictions off?

Yes. The government has said the new protections are designed as defaults rather than fixed blocks, so 16 and 17-year-olds will be able to adjust their own settings if they choose to, unlike the under-16 ban, which removes access altogether.

Why is the government targeting default settings rather than just age verification?

DSIT-commissioned research found that the most commonly encountered age checks, ticking a box or entering a date of birth, are also the ones children rate as least effective, and that 39% of 11 to 17-year-olds have successfully bypassed an age check at least once. Building restrictions into default settings reduces reliance on an age check being passed correctly.

Are AI chatbots covered by these changes too?

The government has said it intends to bring forward separate measures for AI chatbots used by under-18s, including mandatory regular breaks and closer scrutiny of chatbots offering mental health advice, with a ban considered for services judged to pose a serious risk to children. Detail on scope and timing has not yet been published.