Safety alerts and threats

On 8 June the government told Apple, Google and other device makers to switch on technology that detects and blocks nude images on phones and tablets used by children, by default and across the whole device rather than app by app. It gave them three months, so that deadline falls this month. If they have not acted, ministers say they will legislate, with fines for companies and, as a last resort, criminal liability for senior managers. Adults would still be able to see and send such content after an age check. You do not have to wait for any of this. Two similar settings already exist and are worth turning on now: Communication Safety on an iPhone or iPad set up as a Child Account, and Sensitive Content Warnings in Google Messages on a supervised Android phone. Both blur and warn rather than block outright, and neither covers the camera or every app, which is exactly the gap the government is asking the manufacturers to close.

Ofcom has published an update on sites that mainly host harmful material, including suicide and self-harm content, graphic violence and eating disorder posts. After Ofcom contacted them, seven sites blocked UK access altogether (five carrying graphic violence and serious injury material, two dedicated to suicide content) and two more put age checks in place rather than close their doors. None of this needed a formal investigation, which is the point Ofcom is making: quiet pressure is shifting things as well as fines. Nothing to do at home, but it is a reasonable moment to check that your broadband filters and device content settings are still switched on, because sites like these tend to be found through a search or a link shared in a group chat rather than through an app your child already has.

Ofcom has set out how it is policing the Online Safety Act across the 100,000 or so companies covered by it. In the first 18 months it has looked at more than 100 platforms and issued fines totalling over £6 million against 10 providers, with open investigations including TikTok over its protections for children and X over sexualised images generated by its Grok chatbot. The line worth knowing as a parent is the one aimed at the platforms: their existing duties to protect children do not pause or soften because the government plans to ban social media for under-16s. So the age checks, default settings and reporting routes that have appeared over the past year are staying, and are still worth setting up on your child's accounts now rather than waiting to see what the ban looks like.

Roblox has given a legally enforceable undertaking to Australia's online safety regulator, its second notice of concern there in six months. It has three months to deliver on four things: stopping adults contacting children they do not already know without a parent agreeing, making children's accounts private by default, making reporting easier to find and use, and bringing in an independent auditor to check its safety measures, including how it works out a user's age. This is an Australian ruling, but platform changes of this kind usually reach UK players too, and Roblox is one of the apps children mention most when they contact Childline. Nothing changes on your child's account today, so it is worth going into Parental Controls yourself: check the age on the account is right, look at who can chat with them, and have a look at what they are actually playing rather than assuming the age label on a game tells you enough.

The Information Commissioner's Office has published research based on a survey of more than 4,000 children aged 8 to 17 and their parents. It found that 91% of parents use controls or monitoring tools, but 41% of children said they had tried to get around them, and nearly half of those (49%) found it easy. More than a third of children (36%) talk or play online with people they only know online or do not know at all, rising to 40% among 14 to 16-year-olds. Just over half (52%) said they would find a way around an age check if they wanted to use a service. The ICO's point is not that controls are pointless, but that they are a starting point rather than the whole job. Its advice to parents is simple: chat, choose, check. Talk openly about what your child does online, set the controls together, and go back and look again every so often, because what your child uses changes faster than the settings do.

Ofcom has concluded its Online Safety Act investigation into an overseas forum that hosted harmful suicide-related content and has been linked to a number of deaths in the UK. In an update on 17 August, Ofcom confirmed it has now exhausted every legal route available to it. The forum has geoblocked the vast majority of UK users, though Ofcom notes a very small number of already-registered users can still get in. Alongside that, major search engines have removed links to the site, and Ofcom has written to internet providers confirming that net neutrality rules do not stop them filtering suicide-related material. The operator, based in the US, did not pay its £950,000 fine by the June deadline, so Ofcom has begun debt recovery, and it is talking to the government about strengthening its business disruption powers, including court-ordered site blocks. If you are ever worried about your child, talk to them openly and know that free confidential support is available from Papyrus (call 0800 068 4141) and the Samaritans (call 116 123).

For the first time, the NSPCC has published a breakdown of which platforms children bring up when they contact Childline. In the 12 months to March 2026 there were more than 5,500 mentions across the top platforms: TikTok 1,781 (up 37% on the previous year), Snapchat 1,106, Instagram 569, Roblox 550 and YouTube 530. Childline delivered 171,320 counselling sessions to under-18s in that period, around 4,321 of which related to online harms. Some mentions were positive or neutral, but most were negative or involved a child describing something that had happened to them. Almost half of the children who mentioned TikTok were 15 or under. This is a useful prompt rather than a reason to panic: these are the apps your child is most likely to be using, so they are the ones worth going through the safety settings on together, and worth asking about now and then.

Ofcom's annual Media Nations report has found that children aged 4 to 15 are YouTube's largest audience group in the UK, and that YouTube watched on the television has roughly doubled in the past few years. That matters because the TV app is often the version with the fewest controls on it: whatever is signed in on the living room telly is usually a grown-up account with full access. It is worth checking which account YouTube is signed into on your TV and games console, whether Restricted Mode is on, and whether a supervised account or YouTube Kids would suit your younger children better.

Twitch has launched optional parental controls for accounts belonging to 13 to 17-year-olds. A parent creates their own Twitch account, links it to their teenager's, and then gets a dashboard in Account Settings under Family Center. From there you can stop your teenager going live and streaming their own content, turn off Whispers (Twitch's direct messages), set a daily time limit that applies across all their devices, and control whether content carrying certain Content Classification Labels can be recommended to them. Whatever you choose, you also get a weekly email summarising which channels your teenager watched and for how long, and the accounts unlink automatically when they turn 18. Twitch is a big part of gaming culture and has had far fewer parent-facing controls than the consoles do, so this is worth setting up if your child watches streams.

Childnet has followed its age-checks explainer with a complete Roblox guide for parents and carers. It sets out what each account type allows: under-9s (Roblox Kids) have all chat off by default and need permission for higher-rated games, 9 to 12s (Roblox Select) get group chat but not direct chat, and 13-plus accounts have chat on by default though parents can still switch it off up to age 16. It also covers blocking games or players, setting spending limits and turning on spending notifications, and mentions Roblox's new "Family Zone" experience, which walks parents through the controls.

Meta has extended its parent alerts to cover conversations with Meta AI, the chatbot built into Instagram, WhatsApp and Messenger. If a teenager discusses suicide or self-harm with it, a parent using Instagram parental supervision will be notified and pointed to support resources. Meta says the alerts are live in the UK now, it has added a stricter "Limited Content" setting for teen accounts, and it is building a way to alert emergency services if someone appears to be at immediate risk. The alerts only reach you if supervision is already set up, so if your teenager uses Instagram it is worth linking your accounts through Family Centre.

OpenAI has expanded the parental controls available when a parent links their account to a teenager's ChatGPT account. Parents can now set Study Mode as the default, so ChatGPT works through problems with guiding questions rather than simply handing over answers, alongside the existing quiet hours, voice mode and image generation settings. Break reminders now appear more often, and parent alerts have been widened to cover a teen account being suspended for violent content, not just signs of self-harm. If your teenager uses ChatGPT, it is worth linking your account and going through the settings together.

With children at home and online more over the summer, Childnet has put together practical guidance for parents and carers. It covers setting screen-time limits and access windows with parental controls, checking age ratings before your child installs a game or app, asking permission before posting photos of your children, and building enough trust that they will tell you when something goes wrong online. Childnet notes that fear of being judged is one of the main reasons children hide what they do online from their parents.

The government has announced that 16 and 17-year-olds will get an overnight social media curfew from midnight to 6am switched on by default, with autoplay and personalised feeds switched off by default too. Older teenagers will be able to change these settings themselves, so it is a nudge rather than a lock. The measures sit alongside the planned ban on social media for under-16s and are designed to avoid a sudden drop in protection the moment a child turns 16. Regulations are due before Parliament by the end of 2026, with the rules expected to take effect in spring 2027. Nothing changes for your family today, but it is worth knowing what is coming.

Ofcom has published a report saying age checks are now being used at unprecedented scale and are making a difference. The share of children asked to prove their age who met genuinely effective checks rose from 25% to 43% between July 2025 and January 2026, and all of the UK's top 10 porn sites now have age checks. However, Ofcom warns too many services still have weak or no checks, and has opened an investigation into whether TikTok's age checks are effective. Checks now span the porn, social media, dating and gaming sectors. A good moment to make sure your child's age is set correctly on the accounts they use, as this drives what content they can see.

Childnet has released a free set of illustrated "Talk Cards" to help parents and carers start simple conversations about technology with younger children. The cards use pictures and gentle prompts so that talking about being online feels natural, rather than a serious sit-down chat. A handy, no-cost resource if you find it hard to know how to raise the subject with a primary-age child.

Roblox now asks players to complete an age check before they can use chat, either by scanning their face with the camera or, for over-13s, by providing government ID. Players are then placed into a Roblox Kids, Select or 16+ account, and that setting controls who they can chat to and which games they can open. Childnet has published a guide explaining how it works. The important bit for parents: do not sit in front of the camera yourself while your child does the check, because the system will read an adult face and give your child access meant for older players. The age can only be corrected once.

Ofcom has proposed nearly 40 new measures requiring the largest sites and apps, including Google, Bing, Facebook, Instagram, TikTok, Reddit, X, YouTube and Roblox, to do more to stop fraudulent adverts. The rules would require platforms to ban accounts posting scam ads, stop them creating new ones, make scam ads easier to report, and clamp down on criminals using AI to make fake adverts. The proposals are out for consultation until 2 October, with final rules expected next year. It is worth reminding older children and teenagers that adverts promising easy money, free gifts or crypto returns are often scams.

Ofcom has fined the provider of the site fapello.com £600,000 for failing to put highly effective age checks in place, plus a further £30,000 for not responding to a legal information request on time. The site has since geoblocked UK users, so it is no longer directly available here. Ofcom has also opened an investigation into another porn site, eporner.com. This is a further sign that the regulator is enforcing the age-check rules under the Online Safety Act. No direct action is needed from parents, but it is a reminder that device and browser-level content filters remain a sensible backstop.

Internet Matters has published new guidance on what needs to happen for the proposed under-16 social media restrictions to genuinely protect children, drawing on its own family research. It sets out practical points for parents and argues the rules must be backed up with proper support, not just enforcement. Useful reading if you are wondering what the planned changes could mean for your family and how to prepare.