OpenAI Releases New Image Classifier and ChatGPT Warnings Aimed at Combating CSAM
· GlobeWire Daily Editorial Desk · Source: techcrunch.com

Image: Ivan Mehta · Photo source · Resized and letterboxed to 16:9; no AI generation
OpenAI has announced a series of new tools and safety features aimed at combating child sexual abuse material (CSAM). The newly introduced measures include an image classifier for detecting CSAM, warning screens and temporary account blocks in ChatGPT, updated developer guidelines, and specialized investigative tools for law enforcement agencies.
OpenAI stated that its new image classifier can detect visual signs of abuse, including AI-generated imagery. Key details reported about the tool include:
- API Access: OpenAI is providing select companies free access to the classifier via its API.
- Efficiency Metrics: In early testing with partners, the classifier reportedly flagged 48% fewer images while identifying the same volume of abusive content, yielding a higher rate of true positives.
- Collaborations: The system was developed with the National Center for Missing and Exploited Children (NCMEC) and involves cooperation with Tech Coalition and ROOST.
ChatGPT Warnings and Account Actions
To prevent users from attempting to create or access harmful material, OpenAI is updating ChatGPT with new interface measures:
- Alerts and Support: Users attempting to enter prohibited prompts will see warning screens directing them to specialized support resources.
- Temporary Suspensions: Accounts triggering these warnings will face a 24-hour temporary suspension.
- Permanent Bans: Repeated offenses can result in permanent account removal.
The warning messages and behavioral guidelines were reportedly developed in consultation with organizations and researchers, including Protect Children, the Lucy Faithfull Foundation, and researcher Joel Scanlan from the University of Tasmania.
Developer Guidelines and Law Enforcement Support
In addition to consumer-facing features, OpenAI is publishing new guidelines designed to help developers prevent, detect, and report CSAM content using OpenAI tools. Furthermore, the company is equipping law enforcement with AI-powered features, including dedicated plug-ins, investigative tools, and training materials managed by program administrators.
Context and Verification
These feature announcements arrive amid ongoing industry scrutiny regarding AI safety and child protection, following concerns raised by safety advocates and previous lawsuits targeting tech platforms. Observers can verify official platform updates and announcements directly through OpenAI's official corporate communications and security advisory channels.
AI-assisted reporting; source claims are attributed and have not been independently verified.



