OpenAI Reports 80-Fold Spike in Child Exploitation Flags Amid Explosive User Growth

OpenAI sent 80 times as many child exploitation incident reports to a key U.S. clearinghouse during the first half of 2025 as it did in the same period a year earlier, a dramatic increase the company attributes to platform growth and enhanced detection.

According to a recent company update, OpenAI filed 75,027 reports with the National Center for Missing & Exploited Children (NCMEC) CyberTipline between January and June 2025. That number is a staggering leap from the 947 reports it filed in the first half of 2024. U.S. law mandates that companies report apparent child sexual abuse material (CSAM) to the NCMEC, which then forwards credible reports to law enforcement for investigation.

An OpenAI spokesperson, Gaby Raila, said in a statement that the company made investments late in 2024 “to increase [its] capacity to review and action reports in order to keep pace with current and future user growth.” Raila also cited “the introduction of more product surfaces that allowed image uploads and the growing popularity of our products” as contributing factors. This aligns with an August announcement from ChatGPT head Nick Turley, who said the app’s weekly active users had quadrupled year-over-year.

It's important to note that a spike in reports doesn't automatically mean a spike in illegal activity. It can also reflect changes in a platform’s automated moderation or more stringent reporting criteria. In the first half of 2024, OpenAI's 947 reports were about 3,252 pieces of content; in the first half of 2025, the 75,027 reports corresponded to 74,559 pieces of content, indicating a marked rise in both metrics.

This trend is not isolated to OpenAI. NCMEC has observed a broader surge in reports involving generative AI. The center’s own analysis found a 1,325 percent increase in such reports between 2023 and 2024, highlighting a growing challenge across the industry.

OpenAI’s update arrives as the company and its competitors face intense scrutiny over child safety. This past summer, 44 state attorneys general issued a joint letter to major AI firms, warning they would “use every facet of our authority to protect children.” Both OpenAI and Character.AI are also facing lawsuits from families who allege the chatbots contributed to their children’s deaths. This fall, the U.S. Senate Judiciary Committee held a hearing on the harms of AI, and the Federal Trade Commission launched a market study into AI companion bots, which included questions about mitigating risks to children.

In response, OpenAI has rolled out new safety tools. In September, it launched parental controls for ChatGPT, allowing parents to link accounts with their teens and disable features like image generation and voice mode. Following negotiations with the California Department of Justice, OpenAI agreed in October to continue measures to mitigate risks to teens. The next month, it released a “Teen Safety Blueprint,” reaffirming its commitment to detecting and reporting CSAM.

**PRISM Insight:** This staggering increase isn't just a moderation challenge; it's a new financial and ethical stress test for the entire generative AI industry. As models become more powerful and multimodal, the cost of policing user-generated content at scale could become a defining factor in the race for AGI, forcing a difficult conversation about who bears the ultimate responsibility for AI-generated harm.