No single control stops a child using ChatGPT. Qustodio, Bark, Apple Screen Time, Google Family Link and Microsoft Family Safety can each block the app or the website on a device you manage, and two of them can alert you to worrying chats on Android. None covers a school laptop, a friend's phone or a second account. Use them to slow access down, and keep talking.
Last checked November 16, 2026. Every cell in the table below comes from the company's own help or product page, read on that date. I have not tested these apps, and I take no payment or commission from any of them. Where a company's page does not say, the table says "not stated" instead of guessing.
Which parental controls block ChatGPT? The table
Read each row as what the company itself claims for its product, not as a promise.
| Control | Blocks the ChatGPT app? | Blocks chatgpt.com in a browser? | Shows or alerts you to chats? | The gap its own pages show |
|---|---|---|---|---|
| Qustodio (third-party app) | Yes. Its ChatGPT guide says you can "block the ChatGPT app." Its features page lists blocking app by app on Android, and "by category" on iPhone, iPad, Windows, Mac, Chromebook and Kindle | Yes, as a group: "block AI apps and AI websites as an entire category." Which browsers the filter covers is not stated | Android only. Its alerts announcement (September 18, 2026) describes "a real-time alert with a summary of what was discussed" when it detects risky topics. A full transcript is not stated | Alerts are "available for parents protecting Android devices" for now |
| Bark (third-party app, phone and home device) | Yes, in general terms. Its ChatGPT review says "Block ChatGPT entirely, if needed." Which devices: not stated | Its blocking page says "Filter any website, any time." chatgpt.com is not named | Android and the Bark Phone only, text chats only, per its Android monitoring list. It sends "alerts for concerning content." A full transcript is not stated | ChatGPT does not appear on Bark's list for iPhones and iPads |
| Apple Screen Time (built in) | Yes. Apple's guide (published October 26, 2026): Settings > Family > your child > Apps & Websites, pick the app, tap Block App. You can also block new installs or require your approval | Partly. The same screen offers Block Website, and "Ask to Browse" limits a child to approved sites in "Safari and other supported browser apps" | No. Apple's guide describes no way to read or be alerted to chats | Apple devices in your Family group only |
| Google Family Link (Google's own) | Yes, on Android and Chromebook. Google's steps: Screen time > Time limits > App limits, set the app to Blocked | Yes, in Chrome. Google's steps: Controls > Google Chrome and Web > Manage sites > Blocked sites | No | Google says iPhones and iPads cannot be supervised, and on computers a child "may also have access to other browsers where Family Link settings for Chrome won't apply" |
| Microsoft Family Safety (Microsoft's own) | Yes. Microsoft's steps: Apps, features, and games > Block app, on the Windows, Xbox or Mobile tab | Yes, in Edge. Microsoft says its filters "only work in the supported browsers," and other browsers are blocked while the filter is on | No | "You will need to repeat the process for each app, device/platform, and member" |
| ChatGPT's own parental controls | No. They change what a linked teen account can do; they do not block ChatGPT | No | No chat reading. OpenAI's announcement (September 29, 2025) says parents are contacted when there are "signs of acute distress" | They apply to the account you link. OpenAI's help page would not open for me today, so check the live page for changes |
Do Qustodio and Bark block ChatGPT?
Yes, both say they can, and that is the smaller half of what a parent wants to know. Blocking an app or a web address is the oldest job a parental control has. The newer claim is seeing inside the chat, and here the two companies' own pages draw the same line: Android.
Qustodio's ChatGPT alerts review a child's conversations and send a summary when they detect a risky topic. Bark lists ChatGPT under "Chats and responses (text only)" and alerts you to concerning content. Both describe this for Android phones, and Bark adds its own Bark Phone. Neither company's pages document the same thing on an iPhone. If your child has an iPhone, assume a paid app gives you blocking and time limits, and no view of the chat.
That is my reading of their documentation on one day. These products change fast, and "not stated" in my table means exactly that. It does not mean "cannot."
What do the built-in controls on iPhone, Android and Windows do?
They block, and they block well enough for a younger child. Apple, Google and Microsoft each let you block the ChatGPT app and the website on the devices they manage. None of them reads a chat or sends an alert about one.
Each has one catch worth knowing before you rely on it:
- Apple. A blocked app is not deleted. Apple says it is "just dimmed on the Home Screen," so your child will see that you blocked it. Exact steps are in how to block AI apps on your child's iPhone.
- Google. Site blocks live in Chrome. On a Windows or Mac computer, a second browser walks around them. Steps for phones are in how to block AI apps on Android.
- Microsoft. Web filtering works in Edge and blocks other browsers while it is on, and you repeat every app block for each device and each child.
Can any parental control show me what my child typed into ChatGPT?
Not in full, on anything I could find documented. Qustodio and Bark describe alerts on Android when something looks risky. ChatGPT's own controls, once you link your account to your teen's, as OpenAI announced them in September 2025, let you set quiet hours and switch off voice, memory and image generation, and OpenAI says it will contact you about signs of acute distress. Reading the conversation itself is a different matter, and what parents can and cannot see in ChatGPT history covers it. The setup steps for OpenAI's controls are in ChatGPT parental controls.
One thing an alert is not. If an alert, or anything your child says, suggests they may be thinking about harming themselves, a setting is not the response. Talk to them, and contact your pediatrician or GP and the school counselor or safeguarding lead. In the US you can call or text 988. In the UK, Samaritans is on 116 123 and Childline on 0800 1111. A chatbot is not a clinician, and neither is a monitoring app.
Where does every control stop working?
In the same five places, whichever product you buy:
- The school laptop. A school-issued device is managed by the school, and your home controls do not reach it. See what you can do about a school Chromebook.
- Someone else's device. A friend's phone, a cousin's tablet, the library computer.
- A browser you did not filter. Block the app and leave the website open, and you have closed the front door only.
- A second account. ChatGPT's own controls attach to the account you linked. OpenAI says you are notified if your teen unlinks it. A fresh account is another matter.
- AI inside other apps. Blocking ChatGPT does nothing about the assistants built into search, messaging and social apps. What you can switch off on your child's phone, and what has no off switch lists them.
Internet Matters, the UK online safety organization, puts the general point in seven words on its parental controls page: "But remember, parental controls are not fool-proof."
The research suggests parents overestimate what they can see. In a Family Online Safety Institute survey of 1,000 US and 1,003 Australian parents and the same numbers of children aged 10 to 17, run by Ipsos from March 17 to April 6, 2026 and published in July 2026, 27 percent of parents said their child had used generative AI in the past week. Asked the same question about themselves, 38 percent of children said yes. Pew Research Center found a similar gap among US teens aged 13 to 17: in a survey of 1,458 teens and their parents from September 25 to October 9, 2025, 64 percent of teens said they use AI chatbots, while 51 percent of parents said their teen does and about three in ten were not sure. Neither survey shows that controls cause the gap. They show that a parent's picture is usually smaller than the child's.
What I See in Schools
Schools run heavier filtering than any home will, with staff paid to manage it, and students still reach AI tools: on their own phones, on mobile data, at a friend's house. The pattern I see most often in middle and high schools (secondary schools, in the UK) is that a block moves the use to a place no adult can see, and the students who most need an adult's eye are the first to move.
So this is my view, and it is a view, not a finding:
A control slows a child down. Only a conversation tells you where they were going.
I would still use the controls. Slowing a ten-year-old down is worth doing. I would not let the dashboard persuade me that I know what my fourteen-year-old is asking a chatbot.
What should I do tonight, by my child's age?
These are suggestions. Your school's rules and the platform's own age limit come first.
- Under 8. Block the app and the website with the built-in control on the device. No account, no exceptions. If they use AI at all, it is on your lap, on your account.
- 8 to 12. The same block. OpenAI's terms set ChatGPT's minimum age at 13, with a parent's permission up to 18, as our guide to ChatGPT's age rules sets out, so the block is the rule, not your invention. Check the browser as well as the app.
- 13 to 15. A flat block often fails here, because some teachers now expect AI use for schoolwork. Link your account to theirs with ChatGPT's own controls, set quiet hours, and use the device control for time limits. If you add a monitoring app, tell them it is there.
- 16 to 18. Controls matter least and agreement matters most. Write the rules down together; a one-page family AI agreement gives you a starting draft.
What can I say when I turn a control on?
Say it before they find it. A child who discovers a hidden block learns to hide things too. You could say:
"I've blocked ChatGPT on your tablet for now. It isn't a punishment. The company says it's for 13 and up, and I agree with them."
For a teenager:
"I'm not going to read your chats, and the settings don't let me anyway. I've set it to switch off at ten, and if something on there ever worries you, I'd rather hear it from you than from an alert."
And if they have already found a way around it:
"I'm not angry that you got past it. I want to know what you needed it for, because that tells me whether the rule is wrong."
When does the school's rule come first?
On any device the school owns, and on any account the school issued, the school decides what is blocked and what is watched, and you cannot add your own controls on top. Ask the school two questions: which AI tools are open on school devices, and what staff can see. What school monitoring software sees on a school laptop explains the second.
If your child's school has not yet told parents how it handles AI on its own devices, this is the work I do with schools: training staff so the policy means something in the classroom. It is worth sending to your child's principal or head of year. I also write a Sunday newsletter for teachers and parents who want to keep up without the panic.
Dan Fitzpatrick is a former assistant headteacher and Director of Digital Strategy who has trained more than 150,000 educators on AI. Read Dan's story.


