Law enforcement officials are reporting that Meta's artificial intelligence systems are overwhelming child abuse investigators with a deluge of low-quality, unusable tips. This is reportedly draining resources and significantly slowing down investigations into child exploitation cases.
Representatives from the Internet Crimes Against Children (ICAC) Task Force program have stated that Meta's automated systems generate thousands of reports monthly that are often not criminal in nature or lack credible evidence, rendering them unusable for law enforcement.
Benjamin Zwiebel, a special agent with the ICAC taskforce in New Mexico, testified that the volume of these "junk" tips from Meta is substantial. Another ICAC officer noted that their department's cybertips doubled between 2024 and 2025, but the quality has drastically decreased, hindering their ability to take serious action.
Meta has pushed back, asserting its cooperation with law enforcement and highlighting its reporting process, which it claims has been praised by the Department of Justice and the National Center for Missing & Exploited Children (NCMEC). The company stated it resolves over 9,000 emergency requests from U.S. authorities annually, with rapid turnaround times for child safety cases.
However, ICAC officers argue that the increased reliance on AI, potentially exacerbated by staff reductions in human moderation, results in an "overabundance of false positives." This broad net, cast by AI, captures many cases that do not warrant escalation, placing an unsustainable burden on investigators and impacting morale. Meta remains the largest source of reports to NCMEC’s CyberTipline, contributing a significant portion of the millions of tips received annually.