Press release ·
OpenAI adds CSAM detection and ChatGPT bans: schools must teach pupils what AI fakes can do
OpenAI said on 16 November it is releasing a free image classifier for child sexual abuse material and will suspend ChatGPT accounts that seek it. Dan Fitzpatrick says tools help, but schools and parents must still teach children what to do when it happens to them.
In response to: OpenAI releases a new image classifier and ChatGPT warnings to combat CSAM (TechCrunch)
OpenAI announced on 16 November 2026 a set of new tools to fight online child sexual abuse. They include an image classifier that detects child sexual abuse material (CSAM), including AI-generated images, which OpenAI is offering free of charge through an API to selected companies. It was developed with the US National Center for Missing and Exploited Children (NCMEC), and OpenAI is also sharing access through the Tech Coalition and ROOST.
The company says early partner testing showed the classifier flagged 48% fewer images while finding the same number of abusive ones, and that the images it flags are twice as likely to be abusive. It gives reviewers a score rather than a verdict. OpenAI also published guidance for developers on preventing, detecting and reporting CSAM, and is offering tools and training for law enforcement investigators.
For ChatGPT users, there is a new warning. People who try to prompt the app to sexualise children will see an alert pointing them to specialised support, followed by a 24-hour account suspension. Repeat attempts can lead to permanent removal. The app already blocks such requests. OpenAI says it consulted Protect Children and the Lucy Faithfull Foundation on the messaging.
Why it matters beyond the headline: these measures act on the people who make or seek abuse material and on the platforms that host it. They do not reach the child who finds a fake image of themselves circulating in a group chat, or the parent who finds out second-hand. For families and schools, the harm lands in the classroom and the playground, long after any classifier has run.
Dan Fitzpatrick's analysis is that detection is necessary and welcome, and that it is only one layer. A tool that scores images for human reviewers depends on reviewers, reporting routes and adults who know how to respond. A 24-hour suspension changes the behaviour of someone who intends to be caught, not of a pupil experimenting with an app that can do more than they realise.
What leaders and parents should do now: school leaders should check that the safeguarding policy names AI-generated images explicitly and that every member of staff knows who to tell and what to do in the first hour. Teachers should talk to pupils plainly about what AI image tools can fake, and that a child who reports a fake is never in trouble. Parents should ask their child's school how it would respond, and agree at home that telling an adult is always the right move.
Dan Fitzpatrick works with school leaders and teachers on adopting AI safely, with pupils' wellbeing at the centre of every decision, and writes and speaks regularly on what AI means for children and learning.
Dan Fitzpatrick is available for interview and comment on this story. He is available today for live or pre-recorded radio, TV, podcast and print interviews, with a broadcast-quality remote setup for live or pre-recorded down-the-line interviews. Contact: dan@theaieducator.io
“Tools that catch abuse images are welcome, but no classifier replaces a trusted adult a child can tell. Technology firms can stop the worst content spreading, yet they cannot teach a twelve-year-old what to do when a fake picture of them appears in a group chat. Schools and parents have to do that, calmly and before it happens. Curiosity over fear applies here too: talk to children early, and make it safe for them to speak up.”
About Dan Fitzpatrick
Dan Fitzpatrick is an internationally recognised keynote speaker, five-times bestselling author and Forbes contributor, leading the charge on safe and innovative AI adoption. An educator and former senior leader, Dan advises schools, governments and organisations around the world, working with leaders across the United States, UK, Middle East and beyond.
Dan is available for interview and comment. Media contact: Dan Fitzpatrick, dan@theaieducator.io.
Expert comment
Need a comment today?
Dan responds to media requests the same day where he can.
Contact Dan