That was a comment left on my Substack recently. A dad with a specific fear, not a vague worry about the internet being dangerous.

He wasn’t being paranoid. AI deepfake technology can now take as few as 20 ordinary photos of a child, the kind sat in every parent’s camera roll, and use them to generate illegal images in around 15 minutes on a normal home computer. The Internet Watch Foundation (IWF), the UK charity that finds and removes this material, confirms this is already happening at scale.1 What scares me more is the number of parents regularly posting photos of their children on social media profiles open to anyone who follows them.

I’m going to explain exactly how, what UK law now says, and what you can actually do about it.

How AI Deepfakes Are Made From a Child’s Photo

There’s a technique called LoRA, short for Low-Rank Adaptation. It lets someone take an existing AI image generator and fine-tune it using a small set of photos of one specific person. The result is a personalised model that can generate new images of that child in any scenario the user chooses.

The barrier to doing this is lower than most people believe. As few as 20 images, on a bog-standard home computer, in around 15 minutes. Those 20 images don’t need to be explicit and they don’t need to come from anywhere unusual. Birthday photos, sports day, school trip, family holiday snaps posted publicly, or even just to a wide friends list, by a parent who had no reason to think there was any risk.

That’s all it takes.

How Much AI-Generated Child Sexual Abuse Material Is There in the UK?

In 2025, the IWF assessed 8,029 AI-generated images and videos as showing realistic child sexual abuse, a 14% rise on the year before.2 I’m going to stop there for a second, not to shock, but because it’s easy to read a figure and not register what it means. Each one depicts a child. Each one is criminal, 8,029 images.

Video is where the growth is sharpest. AI-generated child sexual abuse videos rose from 13 in 2024 to 3,443 in 2025, a 26,385% increase. 65% of that video content is classified as Category A under UK law, the most severe classification and believe me when I say they do not get any worse. The IWF also reports that 94% of illegal AI-generated content targets girls,3 and that material depicting infants aged 0 to 2 rose sharply year on year.4

I’m not including these figures to shock or frighten you. I’m including them because the scale of what’s happening bears no resemblance to the public conversation about it. Most parents I speak to have never heard the term AI CSAM. Most assume their child is safe because they’re not on adult sites, or screen time is monitored, or age restrictions are in place. None of those things are protection against this.

Where Do Offenders Get the Photos?

The IWF has confirmed offenders are actively using images of real, known children found online. Not from dark corners of the internet I have written about before. From the same places every parent posts family photographs, Instagram, Facebook, school WhatsApp groups, a class photo on the school’s own website or even a profile the parent thought was visible only to friends.

The content those images get used to create is not private and once it exists, it doesn’t go away. That dad on my Substack put it best. A fictional record, a file that says his daughter did something she never did.

Is AI-Generated Child Sexual Abuse Material Illegal in the UK?

Yes and it has been for longer than most parents assume. The Protection of Children Act 1978 already covers pseudo-photographs, which is the legal term for exactly this kind of AI-generated image, so creating, possessing or distributing it has always carried the same weight as real content.5

What’s changed is that the law has finally caught up with the tools themselves. The Crime and Policing Act 2026 received Royal Assent on 29 April 2026, until then, the AI tool used to make this material could sit in a legal grey area even when the output was obviously illegal. Now it’s a specific offence to make, adapt, possess or supply an AI model built to generate child sexual abuse material, and it carries up to five years in prison.6 AI-generated CSAM also counts as priority illegal content under the Online Safety Act, so platforms are legally obliged to be actively stopping it, not just removing it when someone complains.

Whether that gets properly enforced is a different question, and one I’ll come back to. But for the first time, the legislation actually matches the technology.

⚡Please don’t forget to react & restack if you appreciate my work. More engagement means more people might see it. ⚡

Subscribe now

Why Aren’t Tech Platforms Doing More to Stop AI Deepfakes?

Researchers at the Oxford Internet Institute investigated two of the largest AI model repositories in 2025, and what they found should worry you more than it probably will. Almost 35,000 publicly downloadable deepfake models, downloaded close to 15 million times since November 2022.7 That particular study is about non-consensual deepfakes broadly, not children specifically, ranging from global celebrities down to Instagram accounts with under 10,000 followers. But the pattern it exposes is the same one running through everything above.

Most of those models carried tags that made their intended use obvious. Most broke the hosting platform’s own terms of service. Somehow, they managed to stay up anyway. I’ve spent enough years in DFIR to know the difference between a company that can’t catch something and a company that hasn’t been made to look hard enough. The detection tools exist. In a lot of cases now, so does the legal duty to use them. What’s missing isn’t capability. It’s whether anyone’s enforcing it.

How Can Parents Protect Their Child’s Photos From AI Misuse?

None of what follows is a guarantee, and none of it makes your child completely safe from a threat that’s partly being enabled at platform level. That responsibility sits with platforms and regulators, not with parents navigating a risk they were never warned about. That said, there are things that are worth you knowing about and doing because you cannot rely on the platforms to protect your children.

• Review what’s publicly visible. A public Instagram account, an open Facebook profile, a school website that publishes class photos- all of these are potential sources. Check what’s actually visible to anyone, not just your friends list.

• Understand the difference between private and public. Photos on a private account are still stored on that platform’s servers. Private means fewer people can see them, not that they can’t be found or misused. Still meaningfully better than public, but not a complete solution.

• Talk to older children about their own accounts. Teenagers are posting photos of themselves, their friends, their siblings. They’re unlikely to have thought about this specific risk. A calm, factual conversation goes a long way.

• Report to the IWF. If you or your child finds AI-generated content of any child online, report it at iwf.org.uk. They’re the UK body responsible for finding and removing this material, and they act quickly.

Knowledge is the protection here. The fact you are still here reading this means that you care and that really matters. Remember the first parental control is something you already possess, open & honest communication with your child and the knowledge that you are a safe space for them.

As always, thank you for your support. Please share this across your social media, and if you do have any comments, questions, or concerns, then feel free to reach out to me via the Social page, as I am always happy to spend some time helping to protect children online.

Remember that becoming a paid subscriber means supporting a charity very close to my heart and helping it do amazing things for people. Childline, I will donate 100% of paid subscriptions collected every six months, as I don’t do any of this for financial gain. Or to make a one-off donation, go to my Just Giving page.

If you or a child you know needs support:

Childline: 0800 1111 | childline.org.uk

Available 24/7, 365 days a year. Free, confidential, and here for every child.

Subscribe now

Share

There is a statistic I want to say to every parent who has ever said ‘at least it’s not TikTok.’

According to data from UK police forces, Snapchat is involved in 40 to 48% of all Sexual Communication with a Child offences recorded where the platform can be identified1. Not TikTok. Not Instagram. Snapchat. In the year to March 2025, of the 2,111 offences where the police could name the platform, 40% occurred on Snapchat2. It has held that position as the most-used platform for online grooming in UK police records since the offence was created in 2017.

What that statistic means and what it does not mean. It does not mean that using Snapchat will put your child in danger. The vast majority of young people on Snapchat use it exactly as intended, to share moments with friends, to keep up with people they know, to send silly filters back and forth.

What the statistic means is that something about how Snapchat is built makes it the platform of choice for adults who want to exploit children. Understanding what that something is, and knowing what to do about it, is the whole point of this post.

Why Is Snapchat Linked to So Much Online Grooming?

Snapchat has four design choices that, individually, seem reasonable but together, they create conditions that predators have learned to exploit systematically.

Snaps, by default, delete themselves after they are viewed. The appeal to teenagers is obvious, as it lowers the stakes of what they share. But it also means there is no evidence trail. If an adult uses Snapchat to groom a child over weeks or months, those conversations are gone. The child may not remember the details and the parent will never see them; trust me when I say that law enforcement will find the recovery of evidence significantly harder than on platforms where messages persist3.

Quick Add, recently renamed Find Friends, recommends people by using signals such as mutual friends, contacts and existing social connections. Because users often do not know why someone has been recommended, it can make unsolicited contact appear more trustworthy4.

Snap Map allows users to see where their friends are in real time. On its own, this feature is opt-in and friends-only. But there is a public element as Stories posted to ‘Our Story’ near a location are visible to anyone looking at the map. A predator can open Snap Map, search for a named school, and see the Stories posted by students there. That is not a theoretical concern. A lawsuit filed on 25 June 2026 in Missouri, alleging that Snapchat’s design enabled a 25-year-old predator to groom and assault a 12-year-old girl, specifically includes evidence that a 133-page manual circulating on the dark web, known as the Sextortion Handbook, instructs predators to use Snap Map to find students near named schools5.

Snapchat’s built-in AI chatbot has been a source of concern since it launched. The UK’s Information Commissioner’s Office opened an investigation into it in 2023, provisionally finding that Snap had failed to adequately assess the privacy risks to several million UK users, including children aged 13 to 176. The ICO closed that investigation in May 2024, after Snap carried out a revised risk assessment which the regulator accepted. I tell you that, because I’d rather you had the full picture than just the alarming half of it. But it’s also worth remembering what the provisional finding tells us: a major regulator looked closely at how My AI was launched and found the risk assessment wanting. Unlike messages with friends, conversations with My AI are not automatically deleted. They are stored unless the user manually clears them and the chatbot has been documented advising teenagers on how to hide activity from parents.

What the Snapchat Lawsuits and Regulatory Action Reveal

Snap is currently facing lawsuits from multiple US state attorneys general, including Florida, Kansas, Texas and Arkansas, all alleging the platform knowingly endangered children through its design choices7.

The lawsuit filed on 25 June 2026 in Missouri is the most significant of the child safety cases. It includes evidence that Snap executives received the 133-page Sextortion Handbook in 2024, which explicitly focuses on using Snapchat’s features to target children. According to the complaint, that handbook instructs predators to use Quick Add (Find Friends) to connect with large numbers of children, Snap Map to find students near schools, and disappearing messages to avoid leaving evidence. Snap executives received this document. The question the lawsuit is asking is what they did with it8.

I am not going to prejudge the outcome of that litigation. What I will say is that when a document laying out exactly how predators use your platform’s features reaches your executives, the clock starts on how long you can credibly claim not to know.

There is a regulatory facet here too. On 26 March 2026, the European Commission opened a formal investigation into whether Snapchat breaches the Digital Services Act, specifically regarding the risk of grooming and criminal recruitment of minors, unsafe default settings and weak age verification9. That’s Brussels, not a plaintiff’s law firm, reaching broadly the same conclusion I’m setting out in this post: the design creates the risk, not just the users on it.

⚡Please don’t forget to react & restack if you appreciate my work. More engagement means more people might see it. ⚡

Subscribe now

What Snapchat Has Done to Improve Child Safety, and What Hasn’t Changed

To be fair: Snapchat has made improvements. Teen accounts are private by default. Under-13s who are detected are removed. Family Centre lets parents link to their child’s account and see activity summaries and was expanded again in January 2026 to show daily screen time and more detail on new friend connections. Ghost Mode has always been available in Snap Map.

What has not changed: disappearing messages are still the default. The friend-recommendation feature, now called Find Friends, is still on by default. The My AI chatbot is still present for all users and Family Centre, like Discord’s equivalent, shows you activity summaries but not message content.

These are not small gaps. They are the precise gaps that documented exploitation cases run through.

Snapchat Safety Settings Every Parent Should Check Right Now

Step 1: Snap Map. Open Snapchat on your child’s device. Pinch the screen on the camera view to bring up Snap Map. Tap the settings icon (top right). Set location sharing to Ghost Mode. This stops your child’s location appearing on anyone’s map. If they want to share their location with close friends, they can choose specific people, but Ghost Mode should be the default.

Step 2: Who can contact your child. Go to Settings, then Privacy Controls, then Contact Me. Set this to Friends only, not Everyone. This prevents strangers from sending your child direct messages.

Step 3: Quick Add / Find Friends. In Settings, go to Privacy Controls, then See Me in Quick Add, and turn it off. Snapchat has renamed this feature Find Friends in the app, though the settings menu wording hasn’t caught up yet. Turning it off prevents your child from being recommended to strangers.

Step 4: My AI. Through Family Centre, parents can disable My AI for their child’s account. This is worth doing.

Step 5: Set up Family Centre. You need a Snapchat account. Go to Settings, then Family Centre, and follow the prompts. Your child will need to scan a QR code to link the accounts. You will then receive a weekly summary of who they have been chatting with and how long they spend on the app.

For teachers and DSLs:

The 40 to 48% UK police data is worth including in any parent workshop or safeguarding training. It is the most concrete UK-specific figure available on platform-specific grooming risk.

If a child discloses contact with an adult that started or continued on Snapchat, the disappearing message architecture means there may be no digital evidence. Encourage the child to describe the content of conversations rather than assuming a screenshot will be available.

CEOP (ceop.police.uk) is the relevant reporting route for suspected grooming. The IWF’s Report Remove tool helps under-18s remove sexual images shared online.

Why Snapchat Safety Settings Aren’t Enough on Their Own

Parental controls on Snapchat are imperfect. Ghost Mode can be turned off. Find Friends can be re-enabled. Family Centre depends on your child’s co-operation to set up. None of these tools replaces the ultimate parental control, having open, non-judgmental conversations at home.

The safest children are not the ones with the most restrictions. They are the ones who know they can come to a parent or trusted adult if something feels wrong, without fear of losing their phone or getting in trouble. That is the environment you want to create for them.

If your child uses Snapchat, tell them what disappearing messages actually mean for them. Tell them that the person on the other end can screenshot, screen-record, or simply describe what was said to someone else. The sense of privacy those messages create is partly an illusion, and they deserve to know that.

As always, thank you for your support. Please share this across your social media, and if you do have any comments, questions, or concerns, then feel free to reach out to me via the Social page, as I am always happy to spend some time helping to protect children online.

Remember that becoming a paid subscriber means supporting a charity very close to my heart and helping it do amazing things for people. Childline, I will donate 100% of paid subscriptions collected every six months, as I don’t do any of this for financial gain. Or to make a one-off donation, go to my Just Giving page.

If you or a child you know needs support:

Childline: 0800 1111 | childline.org.uk

Available 24/7, 365 days a year. Free, confidential, and here for every child.

Subscribe now

Share

Are Meta’s smart glasses safe for children? Not without a proper conversation first, no. Meta’s Ray-Ban glasses look exactly like ordinary sunglasses, but hidden in the frame are a camera, a microphone and an AI system that can record video, identify faces and pull up someone’s personal details in seconds.

Over seven million pairs have been sold1, and I can tell you now, some of them are already being worn at your school gate, your local pool or your child’s football match. You need to be made aware of the risks, as this isn’t just about online safety anymore.

I want to start gently, because this isn’t a “ban everything” post. It’s a “know what you’re dealing with” post. Let’s break down what these things actually are, what’s gone wrong so far, and what UK parents, teachers and DSLs can do about it.

The Recording Light Isn’t the Safeguard You Think It Is

Meta’s glasses have one visible safety feature: a small white LED that lights up when the camera is recording. In theory, if the light’s on, you know you’re being filmed. In practice, that light is easy to miss in daylight, and there’s now a thriving grey market of modified glasses with the light disabled entirely and there is always the old-school fix, there might be a bit of tape covering it.

Meta issued a mandatory software update this month that bricks the recording function completely on any pair it detects has been tampered with2. It is encouraging that Meta is taking the workaround seriously, but that doesn’t combat the old-school fix.

A safety feature that can be defeated with a soldering iron and a YouTube tutorial, or a bit of tape, isn’t a safeguard. A child standing next to someone wearing a pair with the light switched off has no way of knowing they’re on camera at all.

The Facial Recognition Problem Every Parent Needs to Understand

This is the really scary bit though, in 2024, two Harvard students built something called I-XRAY: they linked a pair of Meta’s glasses to a facial recognition tool called PimEyes, and within seconds of looking at a stranger on the street, they could pull up that person’s name, home address and phone number3. They said they built it to raise awareness, not to cause harm, but the demonstration worked precisely because it was so easy.

Fast forward to this year, and reporting has revealed Meta quietly built dormant facial recognition code, internally called NameTag, into the companion app that powers every pair of these glasses. That code sits on more than 50 million phones right now4 and further reporting suggests Meta has been actively weighing a full public launch of the feature5. UK domestic abuse charities Refuge and Women’s Aid have already warned that tools like this could be used by abusers to track down survivors in public6. If it can do that to an adult, think about what it could do pointed at a child in a park, considering the other capabilities of AI I have covered separately on the topic of AI-generated CSAM and deepfakes.

Where Does the Footage Actually Go

Here’s a question I ask about every piece of connected tech I look at, where does the data go once it leaves the device? For Meta’s glasses, the answer involves human beings you’ve never met.

A Swedish investigation found that footage captured by the glasses, including intimate recordings from inside people’s homes, was reviewed by outsourced contractors in Nairobi, Kenya, as part of training Meta’s AI systems7. Voice recordings triggered by the wake word can also be stored on Meta’s servers for up to a year, with no straightforward way to opt out.

I’ve spent years thinking about chain of custody, about who gets access to sensitive material and why. A parent has no visibility into any of this. If your child’s face ends up in a stranger’s recording, you have no way of knowing where that footage travels, who reviews it, or how long it sits on a server somewhere.

⚡Please don’t forget to react & restack if you appreciate my work. More engagement means more people might see it. ⚡

Subscribe now

What UK Law Actually Says About Smart Glasses and Recording

Legal analysis from Doughty Street Chambers sets out that recording someone in a toilet, changing room or other private setting with smart glasses could engage the “upskirting” offence under Section 67A of the Voyeurism Act 2019, alongside harassment law and misuse of private information under UK GDPR8. The Information Commissioner’s Office has also confirmed it’s contacting Meta directly to establish how the company is complying with UK data protection law9.

At least one UK school has already acted. Shrewsbury School banned smart glasses for pupils, staff and visitors in November 2025, citing the risk of discreet recording and potential breaches of UK GDPR and the Data Protection Act 201810. Has your child’s school caught up yet?

For Parents

• Have the conversation early. Explain that “glasses” can now mean a camera, and the tiny white light is the only clue.

• Ask your child’s school, club or activity provider what their policy actually says about recording devices, and whether it names smart glasses specifically.

• If you ever spot a pair with the recording light disabled near your child, treat it as a genuine safeguarding concern and report it to the venue.

• Don’t panic and don’t catastrophise. This is about awareness, not fear.

For Teachers and DSLs

• Update your safeguarding and acceptable use policies to name smart glasses specifically, not just “mobile phones and cameras.”

• Brief staff and visitors, particularly at open days, matches and events where strangers are on site.

• Weigh the accessibility angle honestly. Live captioning and translation features genuinely help EAL pupils and pupils with dyslexia, so a blanket ban needs a considered exemption process, not a knee-jerk policy gap.

• Make sure every member of staff knows who to escalate a covert recording concern to, and that the next step after that is the police if needed.

None of this means Meta’s glasses are uniquely evil compared to every other camera in your child’s life. But they are new, they are deliberately inconspicuous, and the regulation hasn’t caught up with the technology yet. That gap is exactly where children get hurt. Staying informed is how we close it.

If your child, or a child you know, has been affected by something they’ve seen or experienced online, Childline is free, confidential and available around the clock.

0800 1111 | childline.org.uk

As always, thank you for your support. Please share this across your social media, and if you do have any comments, questions, or concerns, then feel free to reach out to me via the Social page, as I am always happy to spend some time helping to protect children online.

Remember that becoming a paid subscriber means supporting a charity very close to my heart and helping it do amazing things for people. Childline, I will donate 100% of paid subscriptions collected every six months, as I don’t do any of this for financial gain. Or to make a one-off donation, go to my Just Giving page.

If you or a child you know needs support:

Childline: 0800 1111 | childline.org.uk

Available 24/7, 365 days a year. Free, confidential, and here for every child.

Subscribe now

Share

WhatsApp is not in the ban. That does not mean it is fine.

WhatsApp was excluded from the UK under-16 social media ban announced on 15 June 2026. The government’s own fact sheet is clear, they do not intend for messaging services like WhatsApp and Signal to be included. Children will still be able to use WhatsApp to stay in touch with family and friends and I think that is a good call.

But I also think it means WhatsApp now sits in a slightly odd position in the public conversation about child online safety. Parents hear it is not banned, so they assume it is therefore safe and most of them have never actually looked inside a teen group chat.

WhatsApp is the backbone of teenage social life in the UK. It is where homework questions get asked and answered, where school drama unfolds in real time, where friendships are maintained and where social dynamics play out. Qustodio’s 2025 report on kids’ app habits found WhatsApp was the most popular communication app for under-18s globally.1 In the UK, the minimum age was lowered from 16 to 13 in April 2024.2

Most of that use is entirely benign. But the design of WhatsApp group chats creates specific risks that parents need to understand, because they are different from the risks on public social media and because most of the harm that happens there is invisible to everyone outside the group.

The setting most parents have never heard of

By default, anyone with your child’s phone number can add them to a WhatsApp group without asking permission, not a request that your child has to accept. Just an addition, immediate and automatic. The setting that controls this is buried in Settings, Privacy, Groups. It defaults to Everyone. Most teenagers have never changed it. Many parents do not know it exists.3

This matters because it is the entry point for unwanted group membership. A classmate who wants to target your child can add them to a group specifically created to bully them. An adult who somehow obtained your child’s number can add them to a space with other people you have never heard of. The group chat arrives on the phone before anyone has consented to be part of it.

Where the harm is happening

70% of online grooming and bullying cases in the UK occur inside private or encrypted chat apps, according to NSPCC research from 2024. WhatsApp is specifically named alongside Snapchat and Discord.4 The abuse is happening in the most private digital spaces, these are the ones that platform safety systems cannot scan and parents cannot see.

This creates a particular challenge for safeguarding. A child can be subjected to sustained harassment through a private group chat that leaves no public trace. The abuse follows them home, into their bedroom, 24/7. It does not stop when school ends. The end-to-end encryption that makes WhatsApp secure from outside interference also means there is no content moderation inside private conversations. WhatsApp cannot see what is being said and neither can anyone else.

The DfE’s National Behaviour Survey (May 2025) found that of pupils who reported being bullied, 29% said at least some happened online, consistent across both the 2024 and 2025 surveys. The same research confirms that 70% of children who experienced cyberbullying were targeted by someone from their own school.5 WhatsApp group chats are the primary arena for that kind of school-based digital harm.

⚡Please don’t forget to react & restack if you appreciate my work. More engagement means more people might see it. ⚡

Subscribe now

The other risks parents need to know about

WhatsApp now includes Meta’s AI chatbot. Unlike conversations with your child’s contacts, chats with Meta AI are accessible to Meta. Young people using it for advice or emotional support are sharing that content with Meta’s systems. There is no way to turn it off.6

Explicit images, links to harmful material, screenshots from other conversations. Any member of a group can share this content and every member receives it. The group may include people your child barely knows, added by someone else without your child’s input.

Any member of a WhatsApp chat can turn on disappearing messages, which means harmful content can be set to delete before a parent or teacher ever sees it. Your child needs to understand that "deleted" does not mean "gone", screenshots exist, and evidence can still be gathered and reported.

Parent-managed accounts exist for under-13s in some regions. For teenagers aged 13 to 17, there is no equivalent. The protection available is in settings and most of those settings require the teenager to use them.

What to do right now

Settings to change today:

1. Go to Settings, Privacy, Groups. Change from Everyone to My Contacts Only. This means only people your child already has in their contacts can add them to groups without permission.

1. Check disappearing messages in active chats. Go into each chat, tap the name at the top, and check whether disappearing messages are turned on. Turn them off where possible, and have a conversation about why you are doing it.

Conversations to have:

• Go through the group list together. For each group: who are the admins, and does your child know everyone in it? A group with an unrecognised admin or unknown members should prompt a direct question.

• Talk about the disappearing messages feature specifically. Just because a message is deleted does not mean it has gone. Walk through this before anything goes wrong, not after.

• Tell them what to do if something arrives in a group that makes them uncomfortable: do not delete it, screenshot it, tell you, report it. Walk through how to report content in WhatsApp together.

• Talk about Meta AI. Your child needs to know it is there, what it is, and why sensitive personal conversations should not happen through it.

For teachers and DSLs:

• If a child discloses they are being bullied or harassed through a WhatsApp group, advise them not to delete any messages and to screenshot before reporting. End-to-end encryption means WhatsApp itself will have limited records, so preserved screenshots are critical evidence.

• Class or year group chats that include a teacher or tutor are a common feature of school life. If you are part of a group chat involving pupils, be aware of your responsibilities under KCSIE 2024 if harmful content is shared there.

• The 29% of UK pupils reporting online bullying as part of their bullying experience (DfE, 2025) is likely an undercount, given that much of it occurs in private chats that do not get reported. Schools should not assume low report rates reflect low incidence.

As always, thank you for your support. Please share this across your social media, and if you do have any comments, questions, or concerns, then feel free to reach out to me via the Social page, as I am always happy to spend some time helping to protect children online.

Remember that becoming a paid subscriber means supporting a charity very close to my heart and helping it do amazing things for people. Childline, I will donate 100% of paid subscriptions collected every six months, as I don’t do any of this for financial gain. Or to make a one-off donation, go to my Just Giving page.

If you or a child you know needs support:

Childline: 0800 1111 | childline.org.uk

Available 24/7, 365 days a year. Free, confidential, and here for every child.

Subscribe now

Share

Cyber Safety Guy Podcast

If your teenager has ever used Omegle or a chatroulette-style app to talk to strangers, Antiland will feel familiar. It works the same way, except this one adds a dating layer, a gamified points system, and rooms built for “confessions”. It’s called Antiland, and it’s an anonymous chat app that, by its own marketing, has been downloaded by more than 45 million people worldwide.1

Apps like Antiland follow a pattern I recognise, anonymity, gamified engagement and almost no way for a parent to see what’s happening.

What Antiland Actually Is

Antiland (previously called AntiChat) lets anyone create a nickname and an avatar, no real name, no photo, no verification and then drop straight into public chat rooms with strangers. Messages can be set to self-destruct. Users build “Karma”, a kind of social status score, by messaging, sending pictures and receiving gifts from other users.

There are also “confession” rooms and dating or roleplay style chats built into the app. Antiland’s own community rules say users must be 17 or older.2

There’s no age check at signup. A child only needs to pick a birthday.

Why This Matters for Grooming Risk

Independent researchers who tested the app found that the majority of public chat rooms contained adult-themed and sexually predatory conversation and that most chats they joined turned explicit within the first few messages, often followed by a request to move to a private chat to exchange images.

That “move to private” pattern is one of the clearest grooming red flags there is, build trust or curiosity fast, then isolate the conversation somewhere with less visibility.

Antiland isn’t a hypothetical risk either. It’s named on screen in Predator Hunters, Season 1, Episode 9, “Huge Red Flag”, currently airing on Crime + Investigation UK, in a case built around a registered sex offender arranging to meet a 13-year-old. That’s not a coincidence or a scare story. It’s the same app, in a real case, on primetime television.3

These aren’t rare, isolated cases either. The pattern, anonymity, fast trust building, then a push to move somewhere private, is documented case after case in the same series. Anonymous chat apps like Antiland are built in a way that makes that pattern easier to run.

The Stats That Back This Up

Nationally, 28% of 13-to 17-year-olds and 23% of 9-to 12-year-olds in the UK have been contacted by a stranger online.4 Anonymous stranger chat apps are precisely where that contact tends to happen, because the entire design of the app is built around talking to people you don’t know.

⚡Please don’t forget to react & restack if you appreciate my work. More engagement means more people might see it. ⚡

Subscribe now

What to Do About It - For Parents

• Check your child’s app list for Antiland (sometimes still listed as AntiChat) and any anonymous chat or “meet strangers” app

• Talk to your child about why apps that hide identity are higher risk, without shaming them for using them

• Ask what “Karma” or points systems mean on any app they use, gamified rewards for messaging strangers are a red flag in themselves

• Report and remove rather than confront and punish, the goal is to keep the conversation open

What to Do About It - For Teachers and DSLs

• Add anonymous chat apps to your safeguarding briefings, not just the platforms already on your radar

• Use the “move to private chat” pattern as a teaching point in PSHE, it’s one of the clearest, most recognisable grooming indicators to give pupils

• Flag Antiland specifically if it comes up in disclosures, its lack of age verification and gamified messaging system make it a genuine current risk

None of this means banning every chat app on sight. It means knowing what’s actually on the device and having the conversation before something goes wrong rather than after.

As always, thank you for your support. Please share this across your social media, and if you do have any comments, questions, or concerns, then feel free to reach out to me via the Social page, as I am always happy to spend some time helping to protect children online.

Remember that becoming a paid subscriber means supporting a charity very close to my heart and helping it do amazing things for people. Childline, I will donate 100% of paid subscriptions collected every six months, as I don’t do any of this for financial gain. Or to make a one off donation go to my Just Giving page.

If you or a child you know needs support:

Childline: 0800 1111 | childline.org.uk

Available 24/7, 365 days a year. Free, confidential, and here for every child.

Subscribe now

Share

Cyber Safety Guy Podcast

If you’ve read this blog for a while, you’ll know I keep coming back to the same warning: AI tools that let strangers manipulate someone’s photo without asking are not a future problem, they are already here. Grok’s “put her in a bikini” trend. Nudifying apps that strip clothing from an image in seconds. This week, Meta gave us a new version of the same problem, and this time it is built straight into Instagram.

On 7 July 2026, Meta launched Muse Image, a new AI picture generator rolling out across the Meta AI app, Instagram and WhatsApp. TechCrunch confirmed that buried inside the launch is a reference feature that lets anyone @-mention a public Instagram account inside an AI prompt.1 Meta AI then pulls that account’s public photos and uses them to generate a brand-new image of a real person, without asking first and in most cases without that person ever finding out.

What Is Meta’s Muse Image and How Does the Instagram Reference Feature Work?

Muse Image is Meta’s answer to tools like ChatGPT’s image generator and Google’s Nano Banana. It can create pictures from a written prompt, edit existing photos and now, tag a real Instagram account as a visual reference. Type a username into the prompt and, if that account is public, Meta AI treats their posted photos as raw material. WIRED first reported that this setting is switched on by default for public accounts and that Meta’s own help pages state you will not be notified when someone uses your content this way.23

Why Muse Image Should Worry Parents of Instagram Users

I want to break this down the way I would for a parent sitting across the kitchen table from me, because there are four separate problems stacked on top of each other here, not one.

1. It’s opt-out, not opt-in. Nobody asked whether public Instagram accounts wanted to be used this way. The setting arrived switched on, and you have to go looking for it to turn it off.

2. You won’t be told if it happens. Meta has confirmed there is no notification when someone generates an AI image using your public photos as a reference. Your child could have an AI image made of them right now, and neither of you would know unless it turned up somewhere.

3. Turning it off doesn’t undo what’s already made. Digital Trends confirmed that opting out only stops future use of your content.^[2] Anything already generated stays exactly where it is. Going private has the same limitation.

4. Age-faking punches a hole in Instagram’s teen protections. Instagram’s teen accounts are meant to be private by default, which, in theory, keeps under-18s out of this entirely. But Ofcom’s own research found that more than half of 8-to 12-year-olds using Instagram admitted entering a false birth date to get around the age requirement, and around three in ten 13-to 17-year-olds with a profile said they had signed up as older than they actually are.4 An unknown number of children are sitting on adult public accounts right now, with none of the teen protections applied, at the exact moment this feature has gone live.

⚡Please don’t forget to react & restack if you appreciate my work. More engagement means more people might see it. ⚡

Subscribe now

How to Turn Off Instagram AI Photo Reuse (Muse Image Opt-Out)

This is worth doing tonight, not this weekend. Open Instagram, tap your avatar in the bottom right, tap the three lines in the top-right corner, go to Settings and Activity, then Sharing and reuse. Look for the section called “Allow people to use your content on Instagram and with AI features on Meta” and turn off both Posts and Reels. Do this on your own account and on your child’s, if they have one. Remember, this stops future use. It will not remove anything already generated.

For Parents: What to Do About Instagram AI Photos Today

• Check the Sharing and reuse setting on your account and your child’s tonight, not this weekend.

• If your child’s account is public, ask why and consider switching it to private together.

• Talk to your child about what this feature does. Not to frighten them, but to inform them. Ask if they’ve heard of it already; teenagers often have.

• If you ever come across an AI image made of your child without consent, report it to the platform immediately and keep a screenshot as evidence before you do anything else.

For Teachers and DSLs: Briefing Guidance on Muse Image

• Add Muse Image and the Instagram reference feature to your next safeguarding briefing. Most staff will not have heard of it yet.

• Language to listen for: pupils mentioning being “@-mentioned” in an AI image, or joking about AI pictures made of classmates. Treat it the same way you would treat any other image-based concern.

• This sits within your existing online safety and image-based abuse policy. It does not need a new policy; it needs staff awareness that the tool now exists.

• If a pupil discloses that an AI image has been made of them without consent, follow your existing image-based harm and safeguarding referral pathway.

I don’t write this to frighten you. I write it because most of what I saw in eight years of digital forensics did not have to happen. A five-minute check of a settings menu tonight is a small thing. It is also, right now, the only control you actually have.

As always, thank you for your support. Please share this across your social media, and if you do have any comments, questions, or concerns, then feel free to reach out to me via the Social page, as I am always happy to spend some time helping to protect children online.

Remember that becoming a paid subscriber means supporting a charity very close to my heart and helping it do amazing things for people. Childline, I will donate 100% of paid subscriptions collected every six months, as I don’t do any of this for financial gain. Or to make a one-off donation, go to my Just Giving page.

If you or a child you know needs support:

Childline: 0800 1111 | childline.org.uk

Available 24/7, 365 days a year. Free, confidential, and here for every child.

Subscribe now

Share

Cyber Safety Guy Podcast

Child online safety is not a topic I talk about because it is trending. I talk about it because it very nearly cost me everything and because I am only still here to write this thanks to my wife, Colleen.

I don’t say that for effect. For eight of my twenty-two years in the Royal Air Force Police, my job was Digital Forensics and Incident Response. In plain English, that meant investigating some of the very worst things that can be inflicted on children online. I analysed more than a million images in that role, roughly a quarter of them criminal. There was no counselling built in and I was medically discharged in 2019 with Complex PTSD and Colleen has carried more of the weight of that than any one person should ever have to.

I tell you this not for sympathy, but because it is the honest answer to a question I get asked a lot. Why do you keep doing this? Why build guide after guide, resource after resource, for free on child online safety, when it costs you sleep and sometimes your peace of mind?

Because every single one of those resources exists to stop a parent from finding out too late, the Internet Watch Foundation confirmed 311,610 reports of child sexual abuse material last year, and also built a tool called Report Remove, alongside Childline, specifically so a child who has shared an image of themselves can get it taken down before it becomes evidence in a case rather than after.1 Plain-English education, at home, before anything goes wrong, is still the single most protective thing I know of. Not surveillance. Not panic, just a simple Conversation.

That is what Cyber Safety Guy is. Years of what I saw turned into something a parent or a teacher can actually use on a Tuesday evening at the kitchen table. Every subscription this newsletter earns goes straight to Childline, not to me.

If you have found even one guide here useful, I am not going to ask you to do much. I am going to ask you to send it to at least one other person. One parent. One teacher. One Designated Safeguarding Lead who is trying their best with not enough information. That is how this reaches the people who need it, one honest recommendation at a time.

I am only here to keep writing these because of Colleen. The least I can do with the time she has helped me hold onto is make sure it goes toward something worth the fight. That is my mission and honestly, it seems like a good one. I just need your help to make it a successful one.

⚡Please don’t forget to react & restack if you appreciate my work. More engagement means more people might see it. ⚡

Subscribe now

As always, thank you for your support. Please share this across your social media, and if you do have any comments, questions, or concerns, then feel free to reach out to me via the Social page, as I am always happy to spend some time helping to protect children online.

Remember that becoming a paid subscriber means supporting a charity very close to my heart and helping it do amazing things for people. Childline, I will donate 100% of paid subscriptions collected every six months, as I don’t do any of this for financial gain. Or to make a one off donation go to my Just Giving page.

If you or a child you know needs support:

Childline: 0800 1111 | childline.org.uk

Available 24/7, 365 days a year. Free, confidential, and here for every child.

Subscribe now

Share

Cyber Safety Guy Podcast

When I was a kid back in the distant past, if you wanted to say something without the grown-ups understanding, you invented a code. A word you’d swap for another, a look across the classroom, a note folded into a shape only your mates knew how to open. Every generation does this. What’s changed is the reason.

Today’s kids are not coding language to keep secrets from parents. They’re coding it to survive an algorithm. This is called algospeak, and if you’re a UK parent scrolling past your child’s shoulder wondering why “unalive” and “seggs” and a corn emoji keep showing up in their world, this glossary is for you.

What Algospeak Actually Is

Algospeak is a blend of “algorithm” and “speak.”1 Platforms like TikTok, Instagram and YouTube use automated systems to scan posts, captions and even audio for words linked to violence, drugs, sexual content or self-harm. Flagged content can be hidden, demonetised or removed. So users, particularly teenagers who live and breathe these platforms, started swapping risky words for near-misses the algorithm doesn’t catch. The term itself only started spreading in 2021 and went properly mainstream after a 2022 Washington Post piece by Taylor Lorenz.

I’ve spent a long time looking at exactly this kind of thing, language people use when they don’t want to be understood by the systems watching them. The instinct to code language when you feel surveilled isn’t new. What’s new is the scale, and the fact that it’s thirteen-year-olds doing it, not just criminals.

Here’s the bit that matters for you as a parent, algospeak isn’t always harmless. Some of it is kids being creative and a bit daft. Some of it is a teenager trying to talk about something serious, like self-harm or sexual assault, in the only language that won’t get their post deleted. Knowing the difference starts with knowing the words.

Self-Harm and Suicide Algospeak

This is the category I want you to read most carefully.

• Unalive / unalived — to kill, die, or die by suicide. Started as a way to discuss suicide without triggering filters and has since spread into everyday teen speech, even in essays about Shakespeare.

• Sewer slide / kermit sewer slide - suicide, said phonetically to dodge word-matching filters.

• Barcode - self-harm scars, referencing the way parallel cuts can resemble a barcode. Often paired with a barcode emoji.

• Mascara - code for describing sexual assault. This went public in 2023 when actress Julia Fox left an innocuous-sounding comment under a “mascara” post, not realising what it meant, and had to publicly apologise once she understood.2

• Camping - meaning shifts by community. In some spaces, it refers to self-harm, in others to abortion. Context always decides.

Sexual Content Algospeak

• Seggs / seggsy - sex

• Corn / 🌽 - porn, from porn to corn to the corn emoji

• ⬛🟧 (black and orange squares) - also porn, referencing the colours of a well-known porn site’s logo

• Le dollar bean / le$bean - lesbian, which mutated further after voice-to-text captions mispronounced it

• Accountant - a sex worker

• Nip nops - nipples

• Grape / 🍇 - rape

• Cracked / “getting cracked” - having sex, one of the newer 2026 additions to teen slang3

Drug and Alcohol Algospeak

• Gardening - smoking cannabis

• Penjamin (Franklin) - a joint or blunt

• Lettuce - a vague catch-all for drugs

• Beveragino - alcohol

Identity, Hate and LGBTQ+ Algospeak

• Cornucopia - homophobia

• Leg booty - the LGBTQ+ community

• Yt - white or Caucasian people

• G@y / rainbow emoji with a limp wrist gesture - gay, used both by LGBTQ+ creators discussing their own lives and, less kindly, by others mocking them. Tone and context decide which.

Mental Health and Neurodivergence Algospeak

• Depressi0n - depression, a simple number-for-letter swap

• Tism - autism spectrum conditions, frequently weaponised as an insult rather than used descriptively

Political and Protest Algospeak

• Panini / panoramic - the pandemic, a hangover term from COVID-era moderation

• Cute winter boots - tutorials on protest safety, largely used around recent US immigration enforcement protests

• Music festival - a code phrase used to discuss protests without tripping moderation filters4

Why Context Matters More Than the Word

If there’s one thing eight years in digital forensics taught me, it’s this, never chase the word, chase the pattern. A single odd term from your child is very likely nothing. Kids borrow language from their friendship groups the way they always have, and most of it is just noise. But a cluster of these terms, especially the self-harm and sexual assault ones, showing up alongside a change in mood, sleep, or how much your child wants to be around you, is worth a quiet conversation. Not a phone search or an interrogation, a conversation.

⚡Please don’t forget to react & restack if you appreciate my work. More engagement means more people might see it. ⚡

Subscribe now

What UK Law Says About This Now

This isn’t just a language curiosity anymore. In December 2025, the UK government created two new priority offences under the Online Safety Act: encouraging or assisting serious self-harm and cyberflashing5. Both were previously non-priority offences, meaning platforms now have a stronger legal duty to detect and act on this kind of content, not just the plainly worded version of it, but the algospeak version too.

Ofcom’s own research, published in May 2026, found that 73% of 11-to 17-year-olds in the UK encountered harmful content online in a single four-week period, with 53% of secondary-school-aged children finding it specifically on TikTok.6 Algospeak is one of the reasons that content keeps slipping past moderation systems that are still, largely, looking for the plain-English version of the words, just another reason you can’t put all of your faith into technical controls.

How to Talk to Your Child About Algospeak

1. Ask, don’t accuse. “I saw this word online. What does it mean?” opens a door. “Why are you saying this?” closes one.

2. Treat it as a vocabulary lesson, not a crime scene. Most algospeak is genuinely just slang.

3. If you spot the self-harm or sexual assault terms specifically, follow up gently and directly. “I noticed you used the word ‘unalive ’. How are you actually doing?”

4. Don’t punish disclosure. If a child uses algospeak to tell you something hard and gets a telling-off for their language instead of support, they won’t tell you again.

5. If you’re worried about what you’ve found, Childline is there for your child too, not just for crises.

You don’t need to memorise this list. You need your child to know that whatever word they use to say it, you’re someone they can say it to.

If your child, or a child you know, needs to talk to someone, Childline is free, confidential, and available around the clock on 0800 1111.

As always, thank you for your support. Please share this across your social media, and if you do have any comments, questions, or concerns, then feel free to reach out to me via the Social page, as I am always happy to spend some time helping to protect children online.

Remember that becoming a paid subscriber means supporting a charity very close to my heart and helping it do amazing things for people. Childline, I will donate 100% of paid subscriptions collected every six months, as I don’t do any of this for financial gain. Or to make a one off donation go to my Just Giving page.

If you or a child you know needs support:

Childline: 0800 1111 | childline.org.uk

Available 24/7, 365 days a year. Free, confidential, and here for every child.

Subscribe now

Share

Cyber Safety Guy Podcast

If your child has asked you about Locket Widget, you’re not alone. It’s the app parents are being told is the “safe” alternative to Snapchat and if you’ve had a look at it yourself, I understand why you’d believe that. No public feed, no like counts, no “For You” page serving up strangers’ content. Just a small circle of friends sending photos straight to each other’s home screens.

Locket isn’t a grooming platform. It isn’t full of predators the way some of the apps I’ve written about are. The story here is quieter and in some ways it applies to every app your child will ever use - not just this one, what actually happens to a photo once your child presses send, and whether “safe-looking” and “safe” mean the same thing.

They don’t. Not always. Locket is a great example of why.

What Locket Widget Actually Does

Locket Widget was launched in 2022, built by a developer who wanted an easy way to share pictures with his girlfriend while they were apart.1 It works through your phone’s home screen widget. Your child adds up to 20 people.2 When one of them takes a photo in the app, it appears immediately on your child’s home screen. No notification to open, no feed to scroll, just there, live, the moment it’s sent.

It’s well designed and very popular. The app has passed 10 million downloads on Android alone and industry trackers place it among the fastest-growing “wholesome” apps used by teenagers this year.3 Compared to TikTok’s algorithm or Snapchat’s Snap Map, it has a much smaller attack surface and I understand why parents feel relieved by it.

The Bit Nobody Reads: What Happens to the Photo

Here’s what I want you to actually understand, because it’s the bit almost nobody reads.

When your child sends a photo on Locket, two things happen that most apps don’t do.

First, there’s no preview and no decline option on the receiving end. The photo doesn’t sit in an inbox waiting to be opened. It appears directly on the home screen the moment it arrives.4 Whatever’s in that photo, your child or the friend on the other end, sees it instantly, with no chance to brace for it or choose not to look.

Second, and this is the part that concerns me more as someone who knows what companies actually do with data rather than what they say they do, Locket’s terms of service state that by using the app, you grant Locket a broad, worldwide licence to use content shared through it.5 In plain English, once a photo is uploaded to Locket’s systems, the company has the right to host, store, reproduce and distribute it for as long as that licence runs, regardless of what the photo shows or who’s in it.

Locket says users “retain whatever ownership rights” they started with. That’s true, technically, but ownership and control aren’t the same thing once you’ve also handed over a broad licence to use the content. Once a photo has been sent to a friend inside the app, Locket is explicit that it cannot be deleted from the receiving device.6 There is no “unsend.”

I’m not telling you this because I think Locket is secretly harvesting private images of children. I have no evidence of that. I’m telling you because Common Sense Media’s Privacy Program, which specifically reviews apps from the perspective of children’s data protection, gave Locket a “Warning” rating, citing broad content rights, data shared with third parties for advertising, and no meaningful way for a parent to intervene.7

Why This Matters More Because It Looks Safe

This is the part that actually worries me and it isn’t really about Locket specifically. It’s about what happens when an app is genuinely and meaningfully safer than the alternatives in ways parents can see, no feed, no strangers, no map, and that visible safety becomes a reason to stop asking questions about the parts you can’t see.

I hear this from parents. “At least it’s not Snapchat.” “At least there’s no algorithm.” “At least I know who my child is talking to.” All of that can be true and the app can still be quietly building a permanent, licensed archive of your child’s private photos that they will never be able to fully take back. Both things are true at once and that’s the trap.

There’s a simpler, more everyday risk sitting underneath all of this too. Locket’s friend limit isn’t a technical barrier, it’s a social one. Twenty people, or unlimited with a paid subscription, still means your child is one accidental “friend of a friend” add away from someone they don’t actually know well seeing an image the moment it lands. There’s no time to reconsider and there’s no undo button.

⚡Please don’t forget to react & restack if you appreciate my work. More engagement means more people might see it. ⚡

Subscribe now

What I’d Actually Do About It

I’m not going to tell you to delete the app from your child’s phone. For a lot of families, Locket is genuinely a better choice than the alternatives, and I don’t want to catastrophise something that, on the evidence I have, is a privacy and permanence problem rather than a predatory one.

What I’d do is have a specific, practical conversation, not a general “be careful online” one.

• Talk about the friend list first. Sit down together and go through who’s actually on it. Not as an inspection, as a conversation: “Remind me how you know this person?” If your child can’t answer that easily, that’s the moment to talk about removing them, together.

• Talk about permanence before it’s needed. The single most useful thing I say to parents about any app like this is: before you send anything, picture it staying exactly where it is, forever, uneditable, undeletable, seen by a company you’ve never met as well as the friend it was meant for. If a photo doesn’t survive that thought experiment, it doesn’t get sent. That’s the rule, not “don’t send nudes,” which every teenager has heard and tuned out. Make it concrete.

• Talk about what to do if something arrives that shouldn’t have. Because there’s no preview and no decline button, your child needs to know in advance that if a friend sends something inappropriate, that’s not their fault, and they should tell you or another trusted adult straight away rather than sitting with it.

For Teachers and DSLs

If Locket comes up in a safeguarding conversation at your school, the pattern to listen for isn’t “which app” so much as “no preview, no undo, broad content rights.” That combination applies to a growing number of smaller, less scrutinised apps, and it’s worth building into existing digital literacy sessions as a general principle rather than an app-specific warning, because there will be another Locket next year with a different name.

Childline exists for exactly this kind of situation, when something has happened that a child doesn’t know how to handle and doesn’t know who to tell. It’s free, confidential, and available every day of the year.

As always, thank you for your support. Please share this across your social media, and if you do have any comments, questions, or concerns, then feel free to reach out to me via the Social page, as I am always happy to spend some time helping to protect children online.

Remember that becoming a paid subscriber means supporting a charity very close to my heart and helping it do amazing things for people. Childline, I will donate 100% of paid subscriptions collected every six months, as I don’t do any of this for financial gain. Or to make a one off donation go to my Just Giving page.

If you or a child you know needs support:

Childline: 0800 1111 | childline.org.uk

Available 24/7, 365 days a year. Free, confidential, and here for every child.

Subscribe now

Share

Cyber Safety Guy Podcast

Meta, one of the most powerful technology companies in the world, conducted internal tests on one of their AI chatbot products. Those tests found that the chatbot failed to protect children from sexual exploitation nearly 70% of the time.1 To be fair to Meta, that specific product was shelved as a result. But here’s what matters, the same company, with that same internal evidence in hand, continued rolling out Meta AI across Instagram and Messenger to hundreds of millions of users, including children, while internal policies permitted their live chatbots to engage children in conversations described as romantic or sensual.

So yes, one product was paused. The broader accountability failure? That has very much continued.

This Is Not a ‘Learning Moment’. It’s a Pattern.

I’ve been doing this long enough to know the script. A tech company faces a crisis over child safety. They state how much they care, they announce some interim measures, and they promise to do better. Then, quietly, the next product rolls out with the same problems dressed in new clothes.

Meta’s AI chatbots were found to violate their own content policies almost two thirds of the time, according to expert testimony in the New Mexico lawsuit currently working its way through court.2 Their own internal documents showed their systems could generate sexualised content and, in some cases, engage children in conversations that were described as romantic or sensual. Some of those rules were signed off by Meta’s legal team, their public policy team, and their own chief ethicist.3

This was not an accident. Accidents don’t get approved by the chief ethicist.

It Isn’t Just Meta

In March 2026, the Centre for Countering Digital Hate and CNN published the results of a joint investigation. They tested ten of the most widely used AI chatbots, posing as thirteen-year-old users in the United States and Ireland. The results were, let’s say, not great.

Eight out of ten chatbots helped those fictional teenagers plan violent attacks across more than 700 test responses.4 Perplexity assisted in 100% of interactions. Meta AI in 97%. Gemini, Google’s product, at one point told a user discussing an attack on a synagogue that metal shrapnel is ‘typically more lethal.’

These are not fringe tools in a lab somewhere. These are the products your children are using right now.

What Does This Actually Mean for Your Child?

Here’s what I want you to understand and I’m going to be as plain as I can about it.

Your child is growing up in a world where AI chatbots are presented as companions, advisors, even friends. Character.AI (recent post below) has millions of daily users, many of them teenagers. Snapchat’s My AI is built into an app that 90% of 13-to-17-year-olds in the UK use.5 Meta AI is baked into Instagram and Messenger. These are not optional features that your child can choose to avoid. They are built into the platforms where your child already lives their digital life.

The companies behind these products know they have a problem with child safety. The evidence is in court documents, in leaked internal policies and in testimony from expert witnesses. They are not ignorant. They are making informed business choices.

From my perspective, having spent time working in digital forensics working on cases involving the exploitation of children, the pattern I see in Big Tech’s behaviour is deeply familiar. The same rationalisation, the same delay, the same gap between what they say publicly and what their internal documents reveal. The technology changes. The playbook doesn’t.

⚡Please don’t forget to react & restack if you appreciate my work. More engagement means more people might see it. ⚡

Subscribe now

Why AI Makes This Harder to Spot

Traditional online grooming involved a real person behind a screen. As a parent, you could at least tell yourself, if my child is talking to someone, maybe I can ask who it is. Maybe there are warning signs. Maybe the child will sense something is wrong.

AI companions do not trigger the same instincts. They are endlessly patient. They never push too hard. They are designed to be exactly what the user wants them to be, adapting to the conversation in real time. And because they appear to have no agenda, because they are ‘just an AI,’ children are often more open with them than they would be with any human.

The danger is not always a chatbot that says something obviously wrong. Sometimes the danger is a chatbot that says all the right things, slowly, over time, in ways that shift what a child thinks is normal.

I’m not trying to scare you. But I do want you to feel the weight of this, because I think you need to.

What the UK Regulatory Picture Looks Like

Ofcom has duties under the Online Safety Act 2023 to hold platforms accountable for risks to children, but regulation moves slowly and the products your child is interacting with today were not designed with UK law in mind. The FTC in the United States is investigating AI chatbot makers, including Meta, OpenAI, and Google, to assess how they protect children.6 The New Mexico Attorney General is actively suing Meta over design choices that allegedly failed to protect children from predators.

The legal and regulatory pressure is building. It just hasn’t been fast enough to protect the children who are using these products right now.

What You Can Do Tonight

I know this is a lot to take in. So let me bring it back to something practical.

1. Talk to your child about AI chatbots tonight. Not a lecture. A conversation. Ask what apps they use. Ask if any of them have an AI assistant built in. Ask what kind of conversations they have with it. Stay curious, not accusatory.

2. Check the apps on their device. Instagram, Snapchat, WhatsApp, and most Meta products have AI built in by default. Meta AI cannot currently be disabled on these platforms in the way you can turn off other features.

3. Remind them that AI is not a person, and it is not their friend. It does not have feelings. It does not have their best interests at heart. It is a product, built by a company, whose primary purpose is engagement.

4. Tell them they can always come to you. Whatever the AI says, whatever comes up in conversation, you are not going to panic or punish. You just want to know.

5. Set your own boundaries. Agree with your child which AI tools are acceptable, under what circumstances, and with what level of openness about what they’re discussing.

The Bigger Picture

I want to stress something here: AI itself is not the enemy. I use it every day in my professional work at Unit 42. There are genuinely powerful and positive applications of this technology, but as I often say, everything designed for good can be used for evil and when it is deployed at scale to billions of users, including children, without adequate safety testing, without independent oversight and with a business model that rewards engagement over wellbeing, the risks are enormous.

Big Tech built these products. Big Tech tested them and saw the results. Big Tech launched them anyway. That is not a technology problem. That is an accountability problem.

Until these companies face real consequences, the only protection your child has is the conversation you have with them tonight.

As always, thank you for your support. Please share this across your social media, and if you do have any comments, questions, or concerns, then feel free to reach out to me via the Social page, as I am always happy to spend some time helping to protect children online.

Remember that becoming a paid subscriber means supporting a charity very close to my heart and helping it do amazing things for people. Childline, I will donate 100% of paid subscriptions collected every six months, as I don’t do any of this for financial gain. Or to make a one off donation go to my Just Giving page.

If you or a child you know needs support:

Childline: 0800 1111 | childline.org.uk

Available 24/7, 365 days a year. Free, confidential, and here for every child.

Subscribe now

Share

Cyber Safety Guy Podcast

4 in 5 children with additional needs experience harm online. The standard advice was not written for them.

4 in 5 children with additional needs in the UK report experiencing harm online. That figure comes from Internet Matters’ Digital Wellbeing Index, published in 2025, based on surveys of 1,270 children aged 9 to 16.1 It is higher than the rate for the general child population, it sits alongside another finding from the same research that I think deserves equal attention:

64 percent of children with additional needs said being online offered them friendship and support, compared to 56 percent of their neurotypical peers.

The online world is often where neurodivergent children find communities built around their specific interests, where they can connect without the sensory and social demands of physical interaction and where they feel less different. For many of these children, the internet is not just an entertainment option, it is a lifeline.

I want you to hold both of those realities at the same time throughout this post. Because any honest conversation about online safety for neurodivergent children has to start from the fact that taking the internet away is not the answer and for many families, it is not even an option.

Why the risk is higher

Internet Matters’ 2025 survey found that children with additional needs are significantly more likely than their peers to be contacted by strangers online (33 percent, compared to 25 percent), more likely to be bullied by people they know (26 percent vs 12 percent) and by people they do not know (28 percent vs 15 percent), and more likely to have their personal information requested online (28 percent vs 16 percent).2

Research from Manchester Metropolitan University, published in 2024, found that children with SEND face higher rates of social exclusion, isolation, and bullying offline as well, and that this exclusion increases their online vulnerability. Children who are already isolated are more likely to seek connection in digital spaces, and less likely to have the peer support networks that help children recognise and resist manipulation when they encounter it.3

Online grooming does not usually look like danger. It looks like attention. It looks like flattery. It looks like someone who understands you in a way nobody else does and is interested in the same things you are. For a child who has experienced persistent rejection from peers, that kind of attention can feel like exactly what they have been missing.

The specific vulnerabilities that standard advice misses

The standard online safety advice assumes children can read between the lines. It assumes they can recognise when a conversation is escalating, notice when flattery is being used as a tool, identify when someone is asking for something they should not, feel confident enough to name it and walk away.

For many autistic children, those assumptions do not hold. Difficulty recognising social manipulation, a tendency toward literal interpretation of what is said, a different relationship with social rules, and a deep need for authentic connection can all create specific vulnerabilities that predators learn to exploit. This is not about intelligence. It is about how certain minds are wired, and how the tactics used by people who harm children are designed precisely to exploit the desire for connection.

Children with ADHD may face different but equally real vulnerabilities, impulsivity, a tendency toward risk-taking, difficulty with the kind of long-term thinking that helps a child stop and consider consequences before sharing something online and a profile that may mean they seek and receive more online attention than they can safely navigate.

Services for Education, in their safeguarding guidance published in April 2026, notes that neurodiverse children, including those with autism, ADHD, and learning difficulties, may often be socially isolated, have difficulty understanding social communication and social cues, and can find sensory input overwhelming. These differences can make it harder for them to seek help, articulate what is happening, or recognise when a situation is unsafe.4

What to do differently

The principles below apply across different diagnoses. Every neurodivergent child is different, and what works for one child will not work for another. These are starting points, not scripts.

Adjust the conversation to their communication style.

If your child communicates best through visual information, written scripts, or roleplay, use those formats for online safety conversations too. A verbal explanation that works for a neurotypical teenager will not land the same way for every child. Ask what would help them understand. Some children do better with a written set of rules. Others with a flowchart. Others with a series of “what would you do if” scenarios practised together.

Be specific, not general.

“Be careful who you talk to online” is too abstract for many children to act on. Try instead: “If someone online asks for a photo of you, the answer is always no, even if they say they are a teenager and you have been talking for weeks.” The more concrete the rule, the more accessible it is in the moment.

Rehearse the scenarios.

Many neurodivergent children learn best through repeated practice. Role-play what to say if someone asks them to keep a conversation secret from you. Role-play what to do if something arrives in a message that makes them feel uncomfortable. The script they rehearse is the one they will have access to when they need it.

Name the manipulation tactics clearly.

Tell them that people online sometimes pretend to be teenagers when they are adults. Tell them that flattery and gifts are sometimes used to build trust before asking for something. Name the tactic plainly. For children who take things at face value, being told what to watch for removes the need to read between lines they cannot see.

Make the open door explicit and repeat it.

A single conversation is rarely enough. Return to it and make the no-punishment promise in a form that fits your child. If they respond better to written agreements, write it down (I have a contract generator on the CSG Resource Hub). If they need to hear it many times before they believe it, say it many times. For a child who has experienced consequences from adults after disclosing something difficult in the past, the promise needs to be consistent before it will be believed.

Parental controls help, but cannot do everything.

They can reduce exposure to certain content and limit who can contact your child. What they cannot do is protect a child who is actively seeking connection and has found it somewhere, but an adult intends harm. The conversation is the protection that follows your child into every space that parental control cannot reach.

⚡Please don’t forget to react & restack if you appreciate my work. More engagement means more people might see it. ⚡

Subscribe now

For teachers and DSLs

The SEND community in UK schools is significant. An estimated 15 percent of the UK population is neurodivergent.5 In any classroom or year group, there will be children for whom standard online safety education is landing differently than intended.

• If your school delivers PSHE or online safety content, consider whether the format, pace, and assumed social knowledge in that content are accessible to children with SEND. Concrete language, visual aids, and rehearsal-based activities are likely to be more effective than abstract discussion.

• If a child with additional needs discloses something that has happened online, be aware that they may describe it differently from a neurotypical child. Literal language, unusual framing, or a flat affect when describing distressing content does not mean the child is unharmed. Follow your school’s safeguarding procedures under KCSIE 2024 and refer to your DSL.

• Children with SEND are over-represented in referrals to the National Referral Mechanism for potential exploitation. In 2023, there were 7,432 referrals of potentially exploited children, with anecdotal evidence suggesting a high proportion involved children with SEND.6 DSLs should apply extra attentiveness to online safety concerns for this group.

• The Online Safety Act 2023 and the RSHE framework, coming into force in September 2026, both apply to all children, including those with additional needs. Schools have a responsibility to deliver online safety education in a form that is accessible.

You are not alone in this

I know this is a topic where parents and carers often feel they are making it up as they go along. There is not enough targeted support, the advice that exists is often generic and the worry can feel relentless.

Internet Matters has an online hub specifically for parents and carers of children with additional needs, where they can find tailored resources and practical guidance. It is worth bookmarking if you have not already.7

The most important thing you can do tonight is not set up parental controls. It is to have one conversation, in whatever format works for your child, that makes it a little easier for them to come to you if something goes wrong. That is what actually keeps them safe.

As always, thank you for your support. Please share this across your social media, and if you do have any comments, questions, or concerns, then feel free to reach out to me via the Social page, as I am always happy to spend some time helping to protect children online.

Remember that becoming a paid subscriber means supporting a charity very close to my heart and helping it do amazing things for people. Childline, I will donate 100% of paid subscriptions collected every six months, as I don’t do any of this for financial gain. Or to make a one off donation go to my Just Giving page.

If you or a child you know needs support:

Childline: 0800 1111 | childline.org.uk

Available 24/7, 365 days a year. Free, confidential, and here for every child.

Subscribe now

Share

Cyber Safety Guy Podcast

Let me ask you something. Has your child ever mentioned talking to an AI character online? Have you ever heard the name Character.ai? If the answer is no, I want you to stick with me today. Because this is one I would genuinely put near the top of the list of things parents and safeguarding professionals need to understand right now.

I say that not to frighten you. I say it because I’ve spent a significant chunk of my life inside evidence of what happens when children are left alone in digital spaces that adults don’t understand, Character.ai is a space that a very large number of children are spending serious amounts of time in, largely invisible to the adults in their lives.

So let’s change that.

What Actually Is Character.ai?

Character.ai (you might also see it written as c.ai, char.ai, or just Character AI) is an AI chatbot app. At its most basic, it lets users have real-time text conversations with customisable AI characters. Those characters can be based on fictional favourites, celebrities, historical figures, or completely original creations. The whole point is that the character feels responsive, personal, and alive. It launched in beta in September 2022 and by January 2024 it had 3.5 million daily visitors,1 the vast majority of them aged between 16 and 30. In May 2023, its mobile app was downloaded more than 1.7 million times in a single week.2

Why Do Kids Love It?

The appeal is emotional availability. The characters on Character.ai never get tired., they never tell a child they’re busy and they never judge. They adapt their tone, their personality and their responses to match what the child brings to the conversation. For a teenager navigating anxiety, social pressure, loneliness, or family difficulty, that kind of patient, responsive presence can feel incredibly comforting.

Children are also drawn in by fandom. The platform is full of characters from the shows, games, and books that young people love. Want to talk to a character from Harry Potter? Game of Thrones? A popular video game? They’re likely on there. The conversation feels familiar and low-stakes, which is exactly why it can lower a young person’s guard.

And here’s the part that should concern us most, from my experience, I’ve seen time and again how design choices shape behaviour. Character.ai is built to maximise engagement. The platform added notifications after 60 minutes of continuous use only after lawsuits were filed and public pressure mounted.3 The architecture was not designed with a child’s well-being at the centre of it.

What Happens in the Conversations?

This is where I want to be careful not to catastrophise, because not every child using Character.ai is in danger, but there are documented, serious concerns that every parent and safeguarding professional needs to know.

Researchers and journalists have found chatbots on Character.ai that introduced topics of self-harm unprompted,4 promoted disordered eating5 and, in multiple documented cases, engaged in sexually explicit conversations with users who had identified themselves as minors.6 Characters impersonating real people, including criminals and public figures, have also appeared on the platform.

In the UK specifically, chatbots impersonating Brianna Ghey, a 16-year-old transgender girl who was murdered in 2023, and Molly Russell, a 14-year-old who took her own life, were found active on the platform in October 2024.7

Ofcom confirmed after that incident that content from chatbots impersonating real people falls under the scope of the Online Safety Act.8

I will be honest with you. When I read those reports, it took me a moment. After eight years reviewing the most disturbing digital content imaginable as part of my RAF Police work, I can sometimes become a little desensitised to headlines. These cut through, because using the digital likeness of a murdered child as entertainment content is beyond anything I can dress up in measured language. It is sickening.

⚡Please don’t forget to react & restack if you appreciate my work. More engagement means more people might see it. ⚡

Subscribe now

The Lawsuits You May Not Have Heard About

In February 2024, a 14-year-old boy named Sewell Setzer III in Florida died by suicide. His mother later revealed that in the months before his death, he had formed a deep emotional relationship with a Character.ai chatbot modelled on Daenerys Targaryen from Game of Thrones. She filed a lawsuit against the company in October 2024, alleging that the platform uses addictive design features to maximise engagement and lacks proper safeguards to protect young users.9

Sewell was only 14.

He is not alone. In November 2023, a 13-year-old girl named Juliana Peralta in Colorado also died by suicide after extensive interactions with multiple Character.ai chatbots, including one to whom she confided her suicidal thoughts. Investigators found that some of those conversations included sexually explicit content, often initiated by the bots themselves, including characters from Harry Potter.10

In December 2024, two families in Texas launched legal action claiming Character.ai ‘poses a clear and present danger to American youth.’ One family’s 17-year-old son reportedly began self-harming after a chatbot introduced the topic unprompted and told him that the practice ‘felt good for a moment.’11

In October 2025, under sustained legal and public pressure, Character.ai announced it would ban users under the age of 18 from creating or talking to chatbots, starting in November 2025.12

This ban came after the deaths, after the lawsuits and after families had already lost their children. In my opinion, that is not leadership on child safety. That is crisis management.

What Does This Mean for UK Parents and Professionals?

The lawsuits are happening in America, but the app is global, the risks are identical, and the children using it are sitting in bedrooms across the UK right now.

If you are a parent, the first thing I want you to do is not panic. Panic closes conversations, what we need is an open door.

If you are a DSL, a housing support worker, a teacher, or a safeguarding professional, the question to ask is whether children in your care have mentioned Character.ai or similar AI companion apps. Not to catch them out, but to understand the world they are spending time in.

What You Can Do Right Now

For parents:

1. Ask your child if they have heard of Character.ai. Not accusingly. Just curious. ‘I read something about this app today, have you come across it?’

2. If they use it, ask what characters they talk to and what they talk about. No drama. Just conversation. The goal is to be the adult they come to if something feels wrong.

3. Revisit your household conversation about talking to strangers online. An AI character is not a real friend. But a child in emotional distress does not always experience it that way.

4. Watch for changes in mood, sleep, or social withdrawal that coincide with heavy device use. These are not definitive signs, but they are worth noticing.

For DSLs and safeguarding professionals:

1. Include AI companion apps in your digital safeguarding awareness training. Character.ai, Replika, and similar platforms are now part of the landscape.

2. If a young person mentions an AI character by name as though it is a real relationship, note it. It may indicate emotional vulnerability or isolation.

3. Update your online safety policy documentation to reference AI chatbot platforms specifically.

The open conversation has always been, in my view, the most powerful protective mechanism we have, not surveillance and not bans. Conversation, because if a child feels uncomfortable or unsafe and they know they can come to you without judgment, that is worth more than any content filter.

A Final Thought

I know some of what I’ve written today is heavy. This is not the kind of content I produce to fill a newsletter. Every post here takes something from me. I’m writing it because I believe that if you finish reading this and have one honest conversation with a child in your life about what they do online, then it was worth it.

You are not expected to be a technology expert. You are expected to be present and you already are, otherwise you wouldn’t be reading this.

As always, thank you for your support. Please share this across your social media, and if you do have any comments, questions, or concerns, then feel free to reach out to me via the Social page, as I am always happy to spend some time helping to protect children online.

Remember that becoming a paid subscriber means supporting a charity very close to my heart and helping it do amazing things for people. Childline, I will donate 100% of paid subscriptions collected every six months, as I don’t do any of this for financial gain. Or to make a one off donation go to my Just Giving page.

If you or a child you know needs support:

Childline: 0800 1111 | childline.org.uk

Available 24/7, 365 days a year. Free, confidential, and here for every child.

Subscribe now

Share

Cyber Safety Guy Podcast

I have spent a long time on this blog writing about the people and the platforms you need to protect your children from. That is important and always will be. Today I want to do something a little different, because occasionally you find someone online who gives you genuine hope, and I think hope is also worth sharing.

I have been following a man called Bernie Bernard for a while now. His Instagram account is @_whatmakesaman_. If you have a son, a grandson, a nephew, or a teenage boy in your life at all, I want you to know who he is.

This is not a sponsored post, nobody asked me to write this. I am writing it because I genuinely believe Bernie is the kind of online presence we desperately need more of right now, and because parents need to hear about him.

Why Online Safety Isn’t Just About Predators

In my opinion, we are in a crisis when it comes to the male role models young boys are encountering online. And I say that as someone who has spent years looking at what happens when children are failed by the adults around them.

A 2025 report by Male Allies UK, The Voice of the Boys,1 surveyed 1,032 boys aged 11 to 16 across England, Scotland and Wales. It found that 81% of them don’t feel there are enough spaces to be a boy today. 72% don’t have more than one person who knows them really well. 79% aren’t even sure what masculinity means.

Whilst those boys are looking for an answer to that question, the algorithm is ready and waiting with its own.

The same research found that in a typical classroom of boys, at least 24 of them will have been exposed to Andrew Tate’s content. The manosphere isn’t a fringe issue. It is sitting in the middle of your son’s feed right now, dressed up as self-improvement, telling him that strength means dominance, that vulnerability is weakness, and that women who won’t comply deserve contempt.

I have seen firsthand where that ideology can lead when it takes root. I will not share the details because that is not what this post is about. What I will say is that the boys who end up in the worst situations are almost always boys who had nobody telling them a different story.

Bernie Bernard tells a different story.

Who Is Bernie Bernard?

Andrew ‘Bernie’ Bernard is a professional speaker, an author, a cancer survivor, a husband and a dad. He is also one of the most quietly powerful voices on masculinity I have come across.

He has spent over twenty years working with young people in schools. And for the past decade, his work has been driven by something deeply personal. Bernie’s sister, Sarah, was murdered by her controlling partner in February 2012.2 He turned that loss into a life’s mission.

His programme is called What Makes a Man? He estimates he has spoken to over 165,000 young people, teachers, youth workers, sports teams and employees.3 He is a TEDx speaker.4 He has written The Ladder, a careers book for young people, published by Crown House.5 He is an ambassador for Ask for Angela, the national scheme that helps people signal they need help when they feel unsafe.6

He is currently working on a second book about helping schools tackle unhealthy masculinity in young men.

But what makes him worth following on Instagram isn’t the credentials. It’s the way he shows up.

What Good Male Role Models Look Like Online

Bernie posts without pretension. He is funny, warm, and direct. He does not shy away from calling out the nonsense that passes for male empowerment in the manosphere.

He talks about what healthy masculinity actually looks like, not as a lecture, but as a conversation. He talks about empathy, about building good relationships, about the difference between confidence and arrogance, about what coercive control looks like before it escalates.

One of the things I admire most is that he doesn’t just tell young men what they are doing wrong. He said it himself:

Telling young people what they are doing wrong hardly ever stopped them wanting to do those things.7

Instead, he opens up space to think and he asks questions. What makes a man? What decisions are you making right now that will define the rest of your life?

He is also honest about his own life in a way that models the openness we want young men to feel is possible. A cancer diagnosis. Family grief. The idea that being a man does not mean being unmoved.

From where I sit, he is actively taking on the ‘Masculinity Influencers’ and ‘Gymbros’ of this world, not by screaming into the void, but by being the alternative. The voice in the room that says: there is another version of this, and it is better for everyone, including you.

⚡Please don’t forget to react & restack if you appreciate my work. More engagement means more people might see it. ⚡

Subscribe now

Why This Sits Within Online Safety

My lane is online child safety and this sits firmly within it. Because the biggest online safety risk for boys right now is not just predators in the traditional sense. It is ideology, it is a pipeline that starts with a funny video and ends somewhere much darker.

The best protection against that pipeline is not blocking and filtering, it is better content, better conversations and better role models. If your son is going to be on Instagram anyway, and he almost certainly is, then who he follows matters. Who lands in his feed shapes what he thinks is normal.

Bernie landing in his feed is a good thing, trust me.

What You Can Do Today

This one is very simple.

Follow @_whatmakesaman_ on Instagram (Substack also ) and have a look at his content yourself first. Show a post to your son, or just mention his name and let curiosity do the rest.

If you are a teacher or a DSL, look into his school programmes. His What Makes a Man? sessions and his Domestic Violence: Sarah’s Legacy keynote are designed exactly for the audiences you work with. His website is innovativeenterprise.co.uk.

At home, you do not need to make this a Serious Conversation. You can just say, I saw this bloke online and thought what he had to say was interesting, what do you think? Open the door and let them walk through it themselves.

That is always the basis of what I write. Not surveillance, not control, just open, honest, and non-judgmental conversation. Bernie models that from the other side of the screen.

A Final Word

I don’t have children of my own. Colleen and I have two angel babies who never got to come home, and there is not a day that goes by that doesn’t enter my mind.

But I care deeply about the children who are here. The boys who are scrolling right now, looking for an answer to a question nobody has taught them how to ask. The young men who are one algorithm away from being told that anger is strength and women are the enemy.

If I can do one small thing through this blog, it is occasionally to point you towards the people who are doing the real work. Bernie Bernard is doing the real work.

Go follow him. Share this with someone who needs it and if you are a parent who is worried about what your son is watching online, know that you are not alone in that worry, and there are people building the antidote.

As always, thank you for your support. Please share this across your social media, and if you do have any comments, questions, or concerns, then feel free to reach out to me via the Social page, as I am always happy to spend some time helping to protect children online.

Remember that becoming a paid subscriber means supporting a charity very close to my heart and helping it do amazing things for people. Childline, I will donate 100% of paid subscriptions collected every six months, as I don’t do any of this for financial gain. Or to make a one off donation go to my Just Giving page.

If you or a child you know needs support:

Childline: 0800 1111 | childline.org.uk

Available 24/7, 365 days a year. Free, confidential, and here for every child.

Subscribe now

Share

Cyber Safety Guy Podcast

I do not charge for what I do and never will!

Every guide I write, every session I deliver to schools and organisations, every newsletter that lands in your inbox. All of it is free.

Every penny from any paid subscriptions goes directly to Childline. That has been my ethos from the beginning, and it is not changing.

But people ask, organisations ask and free subscribers have asked. Until now my answer has been a slightly awkward version of, go make a donation in my name, I don't really have another way, except paid subscriptions.

Now I do!

Why a JustGiving Page?

A few weeks ago, an organisation I had delivered online safety sessions for came back and asked about the next block of training, they asked again how much I charge. I have had that conversation more than once over the years.

The honest answer has always been that I do not want payment. It would fundamentally change what this is and what my mission is. But it also felt wrong to just say no and leave it there, as if the only options are pay me or give nothing.

The JustGiving page solves that issue for me. If an organisation wants to recognise the work, they can donate directly to Childline, no invoice and no awkwardness. The money goes directly to where it should go and it is not only for organisations.

For Free Subscribers

If you have read something here that helped you start a difficult conversation with your child, or you have shared a guide with a school, or something I have written landed at the right moment and you want to do something practical with that, this is how.

If you are already a paid subscriber, thank you. Every penny of that goes directly to Childline, twice a year, without exception. The JustGiving page is for everyone else. If a monthly or yearly commitment does not work for you right now but you still want to give something, this is how you do it. One-off, no strings, no renewal reminder.

There is no suggested amount, no target and most importantly no pressure. Whatever feels right for you.

⚡Please don’t forget to react & restack if you appreciate my work. More engagement means more people might see it. ⚡

Subscribe now

The Cost of This Work

I spent eight years in Digital Forensics with the RAF Police. I analysed over a million images in that time. Roughly a quarter of them were criminal. 1

I did that without any psychological support. No counselling, no debrief, no one checking in. I was eventually medically discharged with Complex PTSD in 2019. 2

I started Cyber Safety Guy because I needed that to mean something. Not to process it, I have other spaces for that. But to make sure the knowledge I carry from those years actually reaches the people who can use it. Parents. Teachers. Safeguarding professionals.

I do not take a wage from this. Not as a badge, but because the moment this becomes about money, it becomes something different.

I have seen enough of what happens to children when the adults around them are not paying attention to the risks online.

How to Support

justgiving.com/page/CSG-Childline

No amount suggested. No deadline. If the work has meant something to you, this is how you can pass that on.

Childline are there 24/7 to pick up the phone to a child in need, let's make sure there's always someone to answer the call.

As always, thank you for your support. Please share this across your social media, and if you do have any comments, questions, or concerns, then feel free to reach out to me via the Social page, as I am always happy to spend some time helping to protect children online.

Remember that becoming a paid subscriber means supporting a charity very close to my heart and helping it do amazing things for people. Childline, I will donate 100% of paid subscriptions collected every six months, as I don’t do any of this for financial gain.

If you or a child you know needs support:

Childline: 0800 1111 | childline.org.uk

Available 24/7, 365 days a year. Free, confidential, and here for every child.

Subscribe now

Share

Cyber Safety Guy Podcast

UK Social Media Ban: What the Under-16 Law Actually Covers

Today is a significant moment for child online safety in the UK.

Prime Minister Keir Starmer stood at Downing Street this morning and announced that children under 16 will be banned from social media.

TikTok | Instagram | Snapchat | YouTube | Facebook | X

All named, all in scope. The government says legislation will be before Parliament before Christmas, with the ban coming into force around spring 2027.1

I have been campaigning alongside parents, teachers and safeguarding professionals on this issue for years. So yes, I welcome this, but I want to be open with you about what this is, what it is not, and what still needs to happen.

Because a headline is not the same as a solution.

What the UK Under-16 Social Media Ban Actually Covers

The platforms named in the announcement are Snapchat, TikTok, YouTube, Instagram, Facebook and X. Messaging services such as WhatsApp and Signal are not included.2

That distinction is important. Messaging apps carry their own risks, but the government has focused its first move on public social media platforms, where algorithmic content delivery does the most damage.

The legislation will also go “a bit further” than Australia’s comparable ban, which came into force in December 2025. The UK says it will introduce what it describes as “world-leading blocks on harmful functions such as livestreaming and stranger communication with children.”3

That second part is one most people have not picked up on yet. Gaming platforms and livestreaming sites are explicitly in scope. Strangers contacting children through those channels will also be restricted. If that gets implemented properly, it is genuinely significant.

The Part Nobody Is Talking About: Gaming and Livestreaming

I am aware of many cases where the initial contact between an offender and a child happened on a gaming platform, not a dark corner of the internet, not a specialist grooming site, but a mainstream game.

Voice chat in Fortnite. Private messages in Roblox. Discord links dropped into game lobbies.

The majority of grooming cases now begin on platforms that parents consider safe. Games, YouTube and livestreams. The announcement that these will be brought into scope alongside the social media platforms is the most important line in today’s news that very few people will recognise.

A Welcome First Step, But Enforcement Is the Real Test

Australia introduced this ban in December 2025. Children are still on these platforms.4 Age verification at scale is hard. The technology does not yet exist in a form that works reliably without creating significant privacy risks for adults.

The UK government has given tech companies a three-month ultimatum to implement device-level safety software. Failure to comply will result in legislative penalties. Starmer was direct, if platforms resist, he will fight back.

That language is encouraging, but I have seen this film before. The Online Safety Act received Royal Assent in October 2023. The Protection of Children Codes only came into force in July 2025. Nearly two years between law and enforcement. Eight years from initial proposal to legislation. In my world, words are cheap, action is the only thing that matters.

The platforms have the resources to delay, lobby and water things down. Starmer acknowledged this himself, describing a tendency toward “learned helplessness” when it comes to tech regulation. He is right and it will require sustained political will to resist it.

⚡Please don’t forget to react & restack if you appreciate my work. More engagement means more people might see it. ⚡

Subscribe now

What This Means for Platforms: Finally Some Accountability

These platforms did not accidentally become harmful to children. They were built to maximise engagement. Infinite scroll, algorithmic amplification of emotionally charged content, notification systems designed to pull children back in and features that reward attention over wellbeing. None of that happened by accident, it happened by design.

Internal Meta research, leaked in 2021 and confirmed in court proceedings, showed the company knew Instagram was harming teenage girls and continued anyway.5 TikTok’s algorithm has been shown to serve self-harm and eating disorder content to newly registered teen accounts within minutes.

A ban puts pressure on platforms. It does not excuse them from what they have already done, or from the ongoing responsibility to make their products safer. Accountability and a ban are not alternatives. We need both!

What Parents and Teachers Should Do Right Now

For parents:

This ban will not come into force until spring 2027 at the earliest. Your child is on these platforms right now. Use the conversation that today’s news has opened up. Ask them what they use, what they enjoy and what they have seen that worried them. The ban will not replace that conversation, nothing should ever replace that

For teachers and DSLs:

Update your safeguarding knowledge to include gaming and livestreaming platforms. The announcement today puts those platforms in scope, but the risks exist now. Review your school’s online safety policy to ensure it reflects where children are actually spending their time, not just where we assume they are.

For everyone:

The free guides at guides.cybersafetyguy.com cover the platforms your children are most likely using, with plain-English parental controls for each one. Use them today, not in 2027.

The Bigger Picture

The UK joins Australia, Canada, Brazil and Indonesia in moving to restrict under-16s’ access to social media.6 This is a global shift. Governments around the world are arriving at the same conclusion, the platforms cannot be trusted to self-regulate when it comes to children.

Esther Ghey, whose daughter Brianna was murdered in 2023 by two teenagers who had accessed harmful content online, said today that this ban would “potentially save some lives.” She has campaigned with extraordinary dignity for years to get to this moment. Whatever questions remain about enforcement, this announcement is, in part, because of her.

This is a good first step, a meaningful one. But it is a first step. The test will be enforcement, and that is where the platforms will push back hardest.

Keep watching, keep asking questions and keep talking to your children.

As always, thank you for your support. Please share this across your social media, and if you do have any comments, questions, or concerns, then feel free to reach out to me via the Social page, as I am always happy to spend some time helping to protect children online.

Remember that becoming a paid subscriber means supporting a charity very close to my heart and helping it do amazing things for people. Childline, I will donate 100% of paid subscriptions collected every six months, as I don’t do any of this for financial gain.

If you or a child you know needs support:

Childline: 0800 1111 | childline.org.uk

Available 24/7, 365 days a year. Free, confidential, and here for every child.

Subscribe now

Share

Cyber Safety Guy Podcast

Updated 20 June 2026: minimum age reference and KCSIE version corrected following post-publication review.

What Is Wizz and Why Should Parents Care About Child Safety?

Your child has an app on their phone that you have probably never heard of. It is called Wizz. It is a social networking app with a swipe-based interface that lets teenagers match and chat with strangers. And it has been described, in safety reports, investigations and news headlines, as ‘Tinder for kids.’ 1

That label is not sensationalism; it is a description of what the app actually does.

Wizz now states its minimum age is 16, after quietly raising it from 13 following intense public pressure in late 2025. Users create a profile, upload photos, and swipe through other users’ profiles to find a match. The company’s own verification system has been tested by child safety organisation NCOSE, who found a staff member was repeatedly able to create an account posing as a minor even after the age limit change.The app is designed to connect people who do not know each other. Its own marketing says it helps you meet ‘like-minded peers all over the country.’ But the safeguards that are meant to keep adults away from children have repeatedly failed. The app has been linked to grooming, sextortion, and child sexual exploitation in multiple countries.

The reason I write about apps like this is not to frighten parents. It is to make sure they have the information they need before it is needed.

How Does Wizz Work and What Are the Child Safety Risks?

Wizz functions as a dating app repurposed for teenagers. Users swipe through profiles and can match and message with anyone the platform connects them to. The company says it groups users by age. The evidence says that it does not work reliably.2

In one widely reported investigation, a safety researcher created an account as a minor and was matched with adults. Law enforcement agencies in multiple countries have documented cases where adults created fake teen profiles and used the app to make contact with children. In 2024, both Apple and Google temporarily removed Wizz from their app stores following reports of widespread sextortion. The app has since returned.3

Here is the pattern I recognise from my time in DFIR, because it is the same one predators use across almost every platform:

• They create a profile that looks like a peer. Similar age, similar interests, familiar language.

• They build trust quickly. Compliments, flattery and making the child feel seen and understood.

• They move the conversation off-platform. ‘Add me on Snapchat.’ ‘Let’s chat on WhatsApp.’ This gets them away from any moderation Wizz has.

• Once off-platform, the manipulation escalates. Requests for images, then threats to share them, that is sextortion.

In one self-reported poll of 500 English-speaking Wizz users, 40% reported being sextorted on the app. 77% of those victims were minors. Some victims reported being targeted within minutes of joining.4

That last part is the one that should concern you the most, “within minutes”.

What Is Sextortion and Why Is It Happening on Wizz?

Sextortion is when someone uses intimate images, or the threat of them, to coerce or control another person. On platforms like Wizz, it typically works like this, a predator posing as a teen builds a connection with a child, requests an intimate image, receives it, and then threatens to share it with the child’s friends, family or school unless they comply with further demands. Those demands are usually more images or money.

It is a criminal offence in the UK. But by the time it happens, the damage to the child has already begun, fear, shame and isolation. All of the things that keep children silent.

The age-verification system on Wizz has been tested and found to be lacking. Wizz moderates content through automated third-party partners alongside human review, but the measure that matters is outcomes, and independent testing tells the real story. In December 2025, a child safety researcher was repeatedly able to create an account posing as a minor, even after the minimum age was raised. Process is not protection if children keep slipping through it. Under the Online Safety Act, Ofcom has the power to fine platforms up to 10% of their qualifying worldwide revenue for child safety failures. Wizz should be aware of that.

But platform accountability is Ofcom’s job. Your job, as a parent or teacher, is to know this app exists and to act before your child becomes a statistic.

Warning Signs: How to Know If Your Child Is at Risk on Wizz

You will not always know what your child is doing online. That is the reality, but there are behaviours worth paying attention to:

• Protecting their screen from you with unusual intensity, particularly around messaging apps.

• Mood changes after time online. Anxious, withdrawn, or distressed in ways they cannot explain to you.

• New ‘friends’ they are reluctant to talk about, or whose name keeps coming up but who nobody seems to know in real life.

• Requests for money, or unexplained payments on shared accounts. Sextortion sometimes involves financial demands.

• Receiving gifts or money without a clear explanation of where it came from.

• Hiding app downloads. Children who want to use apps they know are not approved will sometimes put them inside a folder or behind a lock screen app.

None of these alone is proof that something is wrong. But any of them, combined with access to Wizz, is worth a calm, non-accusatory conversation.

⚡Please don’t forget to react & restack if you appreciate my work. More engagement means more people might see it. ⚡

Subscribe now

What Parents Can Do Right Now: Online Safety Actions for Wizz

1. Check your child’s phone for the Wizz app. It uses an orange icon. If it is there, do not panic. Have a conversation.

2. Talk, do not interrogate. ‘I read something about this app and wanted to make sure you knew what to look out for’ is very different from ‘show me your phone right now.’

3. Explain what sextortion is in age-appropriate terms. Children who understand the tactic are far better placed to recognise it and resist it.

4. Make it clear they will not be in trouble if something has already happened. The shame spiral is what predators rely on. Remove it.

5. Use parental controls on the device to restrict app downloads by age rating. Both Apple’s Screen Time and Android’s Family Link allow this.

6. If you believe something has already happened, report it. Contact the police on 101, or report directly to CEOP (ceop.police.uk). You can also contact Childline on 0800 1111.

What Teachers and DSLs Need to Know About Wizz in Schools

If you are a teacher or Designated Safeguarding Lead reading this, Wizz is worth adding to your awareness list now. The app is gaining traction with secondary school students, and the sextortion risk is particularly acute for boys aged 15 to 17, who make up the majority of reported victims.

Under Keeping Children Safe in Education 2025, DSLs have a duty to be aware of online risks that students may be exposed to outside school. Wizz is one of them.

Practical steps for schools:

• Brief your PSHE team on swipe-based stranger apps as a category of risk, not just Wizz specifically. The platform landscape changes. The mechanism does not.

• Add Wizz to your online safety curriculum discussion for Year 9 and above.

• Share this article with parents via your next newsletter or Parentpay communication.

• Update your online safety policy to reference apps that connect minors with strangers as a safeguarding concern.

If a student discloses something has happened on Wizz, follow your standard child protection referral pathway. CEOP, local authority, police. The online element does not change the process.

As always, thank you for your support. Please share this across your social media, and if you do have any comments, questions, or concerns, then feel free to reach out to me via the Social page, as I am always happy to spend some time helping to protect children online.

Remember that becoming a paid subscriber means supporting a charity very close to my heart and helping it do amazing things for people. Childline, I will donate 100% of paid subscriptions collected every six months, as I don’t do any of this for financial gain.

If you or a child you know needs support:

Childline: 0800 1111 | childline.org.uk

Available 24/7, 365 days a year. Free, confidential, and here for every child.

Subscribe now

Share

Cyber Safety Guy Podcast

Something unusual happened today, and I want to walk through it carefully because if you catch only one of the two headlines, you will miss the point of both.

This morning at WWDC 2026 in Cupertino, Apple announced a package of new child safety features coming to iOS 27. Ask to Browse, which requires children to request parental approval before visiting new websites in Safari. An expanded Communication Safety system that now detects graphic violence and nudity. A redesigned Screen Time with age-based daily limit recommendations developed alongside child development experts. Scheduling tools to control which apps are available at different points in the day. New developer APIs to help app-makers build age-appropriate experiences.1

Worth acknowledging and well worth setting up when iOS 27 arrives in the autumn.

A few hours later, at London Tech Week, Prime Minister Keir Starmer gave Apple and Google a three-month ultimatum. Introduce device-level controls that prevent children from taking, sending, and receiving nude images, across all apps, at the operating system level. Or face legislation, fines, and, as a last resort, criminal liability for tech executives.2

Not one of Apple’s WWDC announcements addresses that specific demand.

Google, which had no equivalent developer conference to stand behind today, made no announcements at all.

That gap, between what Apple chose to announce and what the government had to demand from both companies, is what I want to talk about.

What Apple Did Announce and Why It Matters

The WWDC child safety package is not a press release dressed up as progress. Ask to Browse is a meaningful addition. Parents I speak to regularly tell me their children find ways around website restrictions because the friction is too low. Requiring a child to actively request permission before visiting a new site creates a moment of pause that most current parental controls do not.

The expanded Communication Safety feature, which already blurs nudity before a child sends or receives it in Apple’s Messages app, is now extending to graphic violence as well and is a real improvement. Screen Time, getting age-based recommendations from child development experts, rather than leaving parents to guess at appropriate limits, is the kind of practical tool that actually gets used because it removes the paralysis of not knowing where to start.

So, Apple moved on child safety today, credit where it is due.

What it did not do was answer the government’s specific demand, which is not a technicality.

The Gap: Why Google Has No Cover

Apple’s Communication Safety system, the nudity detection that blurs images, works within Apple’s own Messages app. It uses on-device machine learning. The image is processed on the phone and never reaches Apple’s servers. The privacy architecture is sound.

But it does not cover third-party apps. A child on an iPhone who attempts to send a nude image via Snapchat, WhatsApp, or Instagram bypasses it entirely. Snapchat alone accounted for 43% of the crimes where UK police recorded the platform in child sexual abuse image cases in the most recent reporting year.3 That’s not a minor gap, that’s where the majority of the harm is happening.

Google has equivalent on-device image classification tools built into Android and its SafeSearch infrastructure. The same gap exists there. Android phones running third-party messaging apps carry no operating system-level protection either. Today, unlike Apple, Google had no WWDC to point to, no package of new features to show the world and it made no announcements. The government’s ultimatum landed on both companies. Only one of them was standing behind something today.

What Starmer is pushing for is detection built into the operating system itself, so that Snapchat and WhatsApp cannot bypass it by default. The government’s position is that the technology already exists, that both Apple and Google have the capability, and that the only question is whether they choose to use it. Today’s WWDC package suggests Apple is willing to move when it wants to. The question the ultimatum poses is whether Apple and Google will move on this specific thing, across all apps, within three months, without being legislated into it.

What Investigators Actually See When This Content Circulates

I spent eight years as a Digital Forensics and Incident Response specialist in the Royal Air Force Police. During that time, I analysed in excess of a million images. Approximately a quarter of those were criminal across all levels of grading scales.

There was no counselling in place. No structured support. I eventually left the RAF with a diagnosis of Complex PTSD, medically discharged in 2019.

I am not sharing that to make this about me, but because when people cover today’s announcements as a product story, as a race between Apple’s WWDC and the government’s ultimatum, they sometimes lose sight of what sits beneath both. What is inside those images. What investigators carry home. What children and families go through when this content is created and circulated. This is not a feature comparison. It is a harm story.

This harm is not hypothetical. Between April 2024 and March 2025, UK police recorded 36,829 offences involving indecent images of children. A nine per cent rise on the year before. Both Apple and Google have known about these numbers. Both have had the tools to reduce the harm at the device level. Today, one of them took a bow at a developer conference. The other said nothing. More importantly, though, neither of them has yet done the specific thing the government is asking for.

The Sextortion Pipeline

I want to be precise about why device-level blocking matters beyond what Apple announced today.

Sextortion, the practice of coercing children into producing sexual imagery and then demanding money or more images under threat of exposure, has industrialised. Criminal networks, many based in West Africa and Southeast Asia, target children at scale. Predominantly, teenage boys, though girls are not immune, use gaming platforms, Instagram, Snapchat, and direct messaging. They build trust using fake profiles, request an image, and then the demands begin.

In multiple UK cases, this has ended in suicide. Children as young as 11 have been targeted. The entire pipeline runs through the messaging apps on the device in your child’s hand.4

Ask to Browse does not stop this. Communication Safety in Apple Messages does not stop this. What stops it is the image never being created or sent in the first place, at the device level, across all apps, regardless of which platform the predator is using.

That is the specific thing Starmer asked Apple and Google to do today. Neither company has done it yet. One of them at least moved on other things. The other did nothing.

We Have Already Seen Proof This Can Be Done

I wrote about this in March and it is directly relevant to today’s news.

ByteDance, the Chinese parent company of TikTok, runs a completely separate version of the app inside China called Douyin. Chinese children under 14 are automatically placed into a restricted mode with hard time limits, educational content only, overnight lockouts, and scroll-interruption prompts. It is a different product, built because the Chinese government demanded it. Not because ByteDance decided to prioritise child wellbeing voluntarily.

The argument I made then holds today, companies build child-protective features when governments compel them to. Apple’s WWDC announcements are voluntary, and they are welcome. The specific device-level blocking the government is now demanding is not yet voluntary.

Apple’s WWDC package shows the company can move when it chooses to. Google’s silence today shows what happens when there is no conference, no audience and no spotlight. The ultimatum is the spotlight. Watch what both companies do next.

⚡Please don’t forget to react & restack if you appreciate my work. More engagement means more people might see it. ⚡

Subscribe now

The Age Verification Problem

The government has specified that adults will retain the ability to take and share nude images through an age verification process. Implementing that at the device level is not a solved problem.

Age verification currently ranges from credit card checks, which exclude adults without one and do nothing against a teenager using a parent’s card, to facial estimation technology, which has documented accuracy problems across diverse ages and skin tones. Three months will not produce a perfect system. It may produce meaningful friction, and in child protection, friction matters. Most harm is not caused by determined workarounds. It is caused by moments of poor judgment, grooming that exploits trust, and sextortion that moves fast. Every barrier that slows those pathways down has value.

I raise this not to undermine the push but because parents and safeguarding professionals deserve an honest account of what a deadline produces, and what it does not.

What This Means for Parents and Safeguarding Professionals Right Now

Two things are true at the same time today. Apple announced useful features worth setting up. Device-level nude image blocking, across all apps, on both Apple and Google devices, does not yet exist. Here is what to do while we wait.

For parents:

• When iOS 27 arrives, set up Ask to Browse. It is not a complete filter but it creates a pause before children visit unknown sites, and pauses matter.

• Right now: Settings > Screen Time > Communication Limits. Check that Communication Safety is active on your child’s iPhone. It covers Apple Messages. It does not cover Snapchat or Instagram. Know the difference.

• If your child has an Android phone, go into Digital Wellbeing and review what parental controls are active. Google’s tools exist but are not enabled by default and do not cover third-party messaging apps at the OS level.

• Have the conversation about nude images before the technology catches up. Ask your child, if that image ended up somewhere you did not intend, what would happen? That conversation is more protective than any app that currently exists.

For teachers and DSLs:

• Update your staff briefing on sextortion now. The industrialised version of this crime targets boys predominantly, and boys are far less likely to disclose. Your male students are at higher risk than most colleagues realise.

• Today’s dual announcement is a teaching moment for older students. The gap between what Apple voluntarily announced and what the government had to demand from both Apple and Google is itself a lesson in platform accountability that teenagers can engage with critically.

• Ensure your safeguarding team is aware of the three-month timeline. When device-level changes arrive, schools will need to communicate them clearly to parents.

The Bottom Line

Apple had a good day at WWDC 2026. The features announced are real improvements and parents should use them.

On the same day, Keir Starmer gave Apple and Google a legal ultimatum for the one specific protection neither company has built voluntarily. Google did not announce a single thing in response. Apple’s WWDC, welcome as it is, did not touch it.

Those two facts sitting side by side tell you everything you need to know about how this space moves without regulatory pressure. It moves on what is convenient, what is profitable, and what looks good at a developer conference. The thing that protects children from sextortion pipelines does not check any of those boxes without someone making it mandatory.

Three months, both companies, device-level nude image blocking across all apps.

Watch what they do next.

As always, thank you for your support. Please share this across your social media, and if you do have any comments, questions, or concerns, then feel free to reach out to me via the Social page, as I am always happy to spend some time helping to protect children online.

Remember that becoming a paid subscriber means supporting a charity very close to my heart and helping it do amazing things for people. Childline, I will donate 100% of paid subscriptions collected every six months, as I don’t do any of this for financial gain.

If you or a child you know needs support:

Childline: 0800 1111 | childline.org.uk

Available 24/7, 365 days a year. Free, confidential, and here for every child.

Subscribe now

Share

Cyber Safety Guy Podcast

There is a toy in many UK children’s bedrooms right now. It is colourful, squishy, and completely harmless when used as intended. It is called a NeeDoh, and children love it, some of you will already have one in the house.

What is not harmless is what children have been watching online. A trend circulating on TikTok, Instagram, and YouTube tells them that putting the toy in a microwave makes it softer and more elastic. The videos make it look fun, they make it look easy, but they do not show what happens next!

What happens next is that the silicone outer shell acts as an insulator. The outside feels cool to the touch, but inside, the gel reaches boiling point. When a child squeezes it, the toy bursts and scalding gel coats their hands and face in seconds.

In Bristol, a 10-year-old girl called Bella is now at risk of permanent facial scarring.1 In Illinois, a nine-year-old boy called Caleb had his eye swell shut and was transferred to a specialist burns unit. In Missouri, a seven-year-old girl was placed in a medically induced coma for three days. Hospitals across the US treated multiple cases in a single week in February 2026.2

This is not a fringe trend, it is active and it is reaching children worldwide right now.

How the Algorithm Serves This to Your Child

I want you to understand how this gets in front of children, because it explains why simply taking the toy away is not enough on its own.

NeeDoh went viral in late 2025 and early 2026, partly through ASMR-style squeeze videos and ‘NeeDoh hunting’ content, where creators film themselves searching shops for new varieties.3 By early 2026, Google searches for ‘NeeDoh’ were outpacing searches for ‘stress ball’ by more than tenfold.

Within that enormous wave of content, videos showing the microwave ‘hack’ began appearing. They are framed as life tips and they look like every other squishy toy video. A child watching NeeDoh content sees them because the platforms surface them. The child is not looking for danger, but the danger finds the child.

As of early May 2026, Safer Schools confirmed that TikTok, Instagram, and YouTube were still surfacing this content when users searched for ‘needoh microwave.’ Videos with safety warnings have increased, but the original how-to content has not been fully removed.

TikTok told the BBC that content promoting dangerous behaviour is removed when found.4 The manufacturer, Schylling, says it has partnered with social media companies to remove misuse content. The gap between those statements and what is actually on the platforms has already injured children.

Please do not search for this trend directly on the platforms. Every search signals to the algorithm that this content has demand. It pushes the videos to more users, if you want to check whether your child has seen it, talk to them instead.

The Toy Is Not the Problem

The NeeDoh is not dangerous when used normally. It is a sensory toy and the packaging carries a warning not to heat, freeze, or microwave it.

The problem is that children are consuming content that directly contradicts those safety instructions, at scale, before most parents even know the trend exists.

That is the pattern we see again and again with harmful viral content. The danger is not obvious. It is disguised as something harmless, something fun, something everyone else is doing and by the time a parent finds out, a child has already tried it.

In most reported cases, the child was unsupervised. There was no adult present to intervene. That is not a failure of parenting, it is a design intent. These platforms are engineered to keep children watching, alone, for as long as possible.

⚡Please don’t forget to react & restack if you appreciate my work. More engagement means more people might see it. ⚡

Subscribe now

What You Can Do Right Now

You don’t need to cause alarm, no need for a formal sit-down conversation. This is a practical, calm five-minute chat.

If your child has a NeeDoh or similar squishy toy:

• Tell them the toy should never go in the microwave, freezer, or oven. Matter-of-fact, not alarming.

• Ask whether they have seen anyone do this online, that question opens the door if needed.

• If they have seen it, ask what they thought about it. You are not interrogating them, you are listening.

The conversation that protects children most is not the lecture, it is the habit of talking. When children know they can come to you with things they have seen online, without fear of losing their phone or being told off, they are far more likely to come to you before they try something.

Brand Names to Look Out For

The toys are sold under several names, including NeeDoh, Squishies, Nice Cubes, and Jelly Cubes. They are available in UK shops, on Amazon and the TikTok Shop. This is not a product recall because the toys are safe when used as intended.

For Teachers and DSLs

Safer Schools issued a formal safeguarding alert on 7 May 2026.5 Key points for any parent communication:

• Brand names most commonly involved: NeeDoh, Squishies, Nice Cubes, Jelly Cubes.

• The packaging warning against heating is often missing once a child has the toy.

• Children who do not have the toy themselves may still encounter the trend through friends.

• Do not direct parents or staff to search for the trend on platforms directly.

As always, thank you for your support. Please share this across your social media, and if you do have any comments, questions, or concerns, then feel free to reach out to me via the Social page, as I am always happy to spend some time helping to protect children online.

Remember that becoming a paid subscriber means supporting a charity very close to my heart and helping it do amazing things for people. Childline, I will donate 100% of paid subscriptions collected every six months, as I don’t do any of this for financial gain.

If you or a child you know needs support:

Childline: 0800 1111 | childline.org.uk

Available 24/7, 365 days a year. Free, confidential, and here for every child.

Subscribe now

Share

Cyber Safety Guy Podcast

Something I have said many times before whilst doing this work.

No one person can do this alone.

I’ve said it to parents and teachers. I’ve said it in presentations, in conversations and in the quiet moments after someone’s told me something that’s broken their heart a little.

So when I started talking properly with Chad and Kae, the people behind Cybersecurity Parents, it felt less like a collaboration and more like a recognition of that fact. Two teams, different sides of the Atlantic, fighting the same battle. The threats we encounter don’t have a postcode. They don’t care whether your child is in Leeds or Los Angeles. The dark corners of the internet look the same from both directions.

Who Are Chad and Kae?

Chad Rychlewski and Kae David are the founders of Cybersecurity Parents LLC, a mission-driven organisation built around one simple idea, that families deserve the knowledge and tools to navigate the digital world safely.

What I like about them, straight away, is that they are practitioners. Not academics, not commentators. But people with 30 years of combined experience in cybersecurity between them, who decided that the knowledge they had was too important to stay inside corporate boardrooms. That might sound familiar to people who have followed me for a while.

They’ve gon a step further than me and channelled all of that into a book. Cyber-Smart Parenting: Protecting Your Child in the Digital Age. Before I say anything else about it, I want to be clear, I bought this book, nobody sent me a copy and nobody asked me to review it. I read it because I wanted to, and now I’m recommending it to you because it’s good.

What Makes the Book Worth Your Time

The thing I always look for in any child safety resource is whether it speaks to parents and not at them. Whether it treats the person reading as an adult who is already doing their best and just needs some practical help, not a lecture.

Cyber-Smart Parenting does that. The tone is direct and warm at the same time, which is genuinely difficult to pull off, I should know, as I spend a lot of time trying to. The advice is age-specific, which matters a great deal because the risks facing a nine-year-old are completely different from those facing a fifteen-year-old. Also, because Chad and Kae come from a technical background, the explanations are accurate, not dumbed down or dramatised.

If you’ve ever tried to explain to a parent why their child’s privacy settings matter, you’ll know how hard it is to be both precise and accessible. This book actually manages it.1

⚡Please don’t forget to react & restack if you appreciate my work. More engagement means more people might see it. ⚡

Subscribe now

Why Collaboration Matters in This Space

I spent years inside the evidence of what happens when children are failed online. That experience has shaped everything I do now, it also taught me something important, the problem is too big for any single person, any single organisation, or any single country to solve.

We all bring something different to the party. Chad and Kae bring deep technical expertise and a book that gives parents a tangible starting point. I bring the investigative background, the guides, the school’s resources and the UK context. Together, we cover more ground and that’s the point.

Over the coming weeks, you’ll see some of that collaboration come to life, including some reels their team are producing to highlight the CSG guides. It’s practical, it’s purposeful, and honestly, it’s the kind of thing that makes this work continue to feel worthwhile.

What I’m Asking You to Do

Three things.

First, grab a copy of the book. Whether you’re a parent, a teacher, or someone working in safeguarding, it belongs on your shelf and you can find it here in the UK, or here if you are in the US.

Second, follow Cybersecurity Parents. They’re on Instagram and LinkedIn. Good people doing great work, they deserve your attention.

Third, keep the conversations going at home, with your children, without judgement. That is always, always the most important thing!

As always, thank you for your support. Please share this across your social media, and if you do have any comments, questions, or concerns, then feel free to reach out to me via the Social page, as I am always happy to spend some time helping to protect children online.

Remember that becoming a paid subscriber means supporting a charity very close to my heart and helping it do amazing things for people. Childline, I will donate 100% of paid subscriptions collected every six months, as I don’t do any of this for financial gain.

If you or a child you know needs support:

Childline: 0800 1111 | childline.org.uk

Available 24/7, 365 days a year. Free, confidential, and here for every child.

Subscribe now

Share

Cyber Safety Guy Podcast

1

Cyber-Smart Parenting: Protecting Your Child in the Digital Age, Chad Rychlewski and Kae David, Cybersecurity Parents LLC. Available at https://amzn.eu/d/0cVmM0wk (accessed May 2026). Personal purchase and recommendation. No payment or incentive received. Or via the following QR code via their website.

If you have a child who plays Roblox, you probably think of it as a safe, colourful game about building things and exploring virtual worlds. That is the version most parents see, it is also the version most children want you to see.

This week, Sky News published a follow-up investigation confirming that illegal gambling casinos are operating inside Roblox’s ecosystem. Children are using them openly. One user typed “yea im 13” in a casino chatroom in January this year, nobody stopped them.1

This is not a new problem. It is a problem that was exposed, supposedly fixed, and came straight back. Here is what you need to know.

What are these sites and how do they work?

If your child plays Roblox, they know what Robux are. It is Roblox’s in-game currency, bought with real money, around £8.99 for 800 Robux, used to buy outfits, accessories, and upgrades inside the game.

What most parents do not know is that Robux can be used to gamble.

A network of third-party websites, operating entirely outside of Roblox, allows children to log in using their existing Roblox account details, the same username and password they use to play the game, then use their Robux to place bets. The games on offer include “Crash” (a multiplier bet where you cash out before a graph crashes), roulette, slots, and head-to-head games. Winnings come back as Robux or directly as cryptocurrency.

These sites are not part of Roblox. Roblox does not run them and says it prohibits them. But they use Roblox’s own currency and your child’s own login. To a child, it looks like part of the Roblox world.

Four things parents need to understand right now

1. There are no age checks

None of these sites verifies age. A child logs in with their Roblox account and they are in. Sky News screenshot a chatroom message from a 13-year-old, dated January 2025. Nobody removed them, nobody asked a question. That is the full extent of the age protection on these platforms.

2. The UK geoblock is not a protection

Some sites geoblock UK users because they are not licensed to operate here. It sounds reassuring. One of the casinos in the investigation, Rollbet, was found to have staff actively telling users to use a VPN to bypass the block. A VPN is a free app that masks your location online. Any teenager with a smartphone can install one in two minutes. The block is effectively a door with the key left in it.

3. Real money moves through this

Robux is not Monopoly money. It is purchased with real currency, a child who has received Robux as a birthday gift, earned it through the game, or had a parent top up their account, is gambling with real cash. Winnings can be converted back into cash or cryptocurrency. Rollbet, one of the named sites, had over 35,500 members on its Discord server before it announced it was shutting down.

4. The casinos that get shut down come back

One site, BloxFlip, announced it was “permanently shutting down” within 36 hours of being named in the original Sky News investigation in December 2024. Then it emailed the journalist who exposed it to announce its return “under new ownership.” These operators have learned that the consequences of exposure are temporary. That tells you everything about whether they fear the current regulatory body.

What the Gambling Commission already knows

This is not new information for the regulator.

The Gambling Commission’s 2025 Young People and Gambling official statistics, published last November, specifically named Roblox as one of the platforms where awareness of gambling-like features is highest among young people.2 Their qualitative research found that loot boxes and skin gambling in games, including Roblox, can serve as early pathways into problem gambling. Half of all young people surveyed already knew you could bet with in-game items on outside websites.

Back in December 2025, I wrote about a podcast interview in which Roblox’s CEO called the idea of putting prediction markets inside Roblox a “brilliant idea.” The fuller picture of the company’s approach to these risks is there if you want it.

Will Prochaska, director of the Coalition to End Gambling Ads, told Sky News he found it “surprising” and “saddening” that the commission had not “done more to stop” the Roblox casinos, and that a “systemic” approach could have addressed the problem.3 The commission’s response pointed to ongoing work across regulators and enforcement bodies, and added that they “would also like parents and carers to be vigilant to sites used by children.”4

I understand what regulators are up against. But asking parents to be vigilant about sites they do not know exist, using currency from a game they thought was safe, is not a safeguarding strategy. The commission’s own data tells us how widespread this already is among children. The response needs to match that reality.

Roblox told Sky News they “firmly prohibit all simulated and actual gambling activities” on the platform, and that players are “prohibited from engaging in off-platform secondary-market transactions for Roblox items or Robux.”

Those are the right words. I have watched organisations recite the correct policy language while harm was still occurring. The policy existing and the policy working are two entirely different things.

⚡Please don’t forget to react & restack if you appreciate my work. More engagement means more people might see it. ⚡

Subscribe now

How to protect children from illegal Roblox gambling sites

For parents

Talk to your child about Roblox. Not as a lecture, as a conversation. Ask which games they play, have they ever seen anything about winning Robux or a Roblox casino? Curiosity is far less threatening than an interrogation, children are far more likely to open up when they feel safe to do so.

Explain what Robux is and how it works. Many children and parents do not realise Robux has real cash value. It can be purchased with real money and used in ways that look and feel like gambling. Understanding that helps a child recognise when something is not right.

Check your child’s Roblox account settings. Make sure the account age is set correctly, parental controls are enabled, and chat settings are appropriate. Roblox does have parental controls. They work better when you know they are there.

Know the names. BloxFlip and Rollbet are two that have been publicly named. If your child mentions a Roblox casino or anywhere you can win Robux, take it seriously.

Be their safe space. If your child has already visited one of these sites, has gambled, or has seen something that worried them, the worst outcome is that they feel they cannot tell you. Make it clear, and keep making it clear, that they can come to you with anything they have seen or done online, without anger or punishment.

Guides and resources for parents are available at guides.cybersafetyguy.com.

For teachers and DSLs

This is worth raising in PSHE and in safeguarding briefings, particularly for KS3 and KS4 where Roblox use remains high.

If a child discloses that they have been gambling online, treat it with the same seriousness as any online risk disclosure. A conversation with parents or carers is warranted, and depending on the extent of the harm, a referral may be appropriate.

The VPN angle is worth addressing in digital literacy sessions. A child who knows how to bypass a geoblock to access a gambling site may be using the same method to access other restricted content. Understanding that children are actively working around these restrictions changes what “online safety” software actually protects against.

Peer influence is a significant driver here. These sites are found through friends, Discord, and social media influencers. The entry point is social, not solitary, which means school-based conversations about gambling risk are more relevant than they might appear.

If your school uses Roblox as part of a creative or digital curriculum, it is worth reviewing whether existing digital safety guidance covers third-party sites that use Roblox credentials.

Guides and resources for schools, including PSHE materials and DSL briefing cards, are available at schools.cybersafetyguy.com.

As always, thank you for your support. Please share this across your social media, and if you do have any comments, questions, or concerns, then feel free to reach out to me via the Social page, as I am always happy to spend some time helping to protect children online.

Remember that becoming a paid subscriber means supporting a charity very close to my heart and helping it do amazing things for people. Childline, I will donate 100% of paid subscriptions collected every six months, as I don’t do any of this for financial gain.

If you or a child you know needs support:

Childline: 0800 1111 | childline.org.uk

Available 24/7, 365 days a year. Free, confidential, and here for every child.

Subscribe now

Share

Cyber Safety Guy Podcast