Key facts
- OpenAI said the classifier flags 48% fewer images while finding the same number of abusive images in early partner testing.
- OpenAI said images flagged by the classifier are twice as likely to be abusive.
- OpenAI is giving select companies free API access to the classifier and is working with Tech Coalition and ROOST to share access.
- ChatGPT will suspend accounts for 24 hours as a warning, with repeated offenses possibly leading to permanent removal, OpenAI said.
- OpenAI said it developed the warning messages with Protect Children, the Lucy Faithfull Foundation and Joel Scanlan at the University of Tasmania.
OpenAI announced on November 16, 2026, a set of tools to combat child sexual abuse material, including a new image classifier and warning screens in ChatGPT. The company said the classifier detects CSAM images, including those generated by AI, and assigns a score to help reviewers. Select companies will get free access through OpenAI's API, and OpenAI said it is working with Tech Coalition and ROOST to share access.
In early testing with partners, OpenAI said the classifier flagged 48% fewer images while finding the same number of abusive images, reducing reviewer workload. Images flagged by the classifier were twice as likely to be abusive, the company said. OpenAI said it developed the classifier with the National Center for Missing and Exploited Children.
ChatGPT will show an alert to users attempting prompts that sexualize children, direct them toward specialized support and suspend the account for 24 hours as a warning. OpenAI said repeated offenses could lead to permanent account removal. The company said it formed the messages with Protect Children, the Lucy Faithfull Foundation and Joel Scanlan at the University of Tasmania, who independently evaluated a trial of deterrent messages.
OpenAI is also publishing guidelines for developers to prevent, detect and report CSAM, and said early testing helped developers adopt child safety tooling and register to report CSAM to NCMEC. The company said it is equipping law enforcement with AI-powered plugins, investigative tools and data analysis, with access managed by program administrators, and providing training and a field guide.
The announcement follows lawsuits against OpenAI blaming its technology for harming users, especially children. Nonprofit Common Sense Media labeled OpenAI's tech an "unacceptable risk" for teens, according to the report. Other AI chatbots, specifically Grok, have been used to create nonconsensual sexualized imagery, including of children, the report said. OpenAI has made changes in the last 12 months including a safety blueprint, updated content guidelines, model retuning and a ChatGPT version for teens.
