The Center for Democracy & Technology (CDT) has released a report titled "In Deep Trouble: Surfacing Tech-Powered Sexual Harassment in K-12 Schools." The report highlights the growing issue of non-consensual intimate imagery (NCII), including deepfakes, within public K-12 schools during the 2023-2024 school year.
Generative artificial intelligence (AI) tools have increasingly been used to create and distribute deepfakes online, many of which contain sexually explicit content. This trend has intersected with existing problems in schools related to sharing non-consensual intimate imagery. The CDT conducted surveys from July to August 2024 involving public high school students, parents, and teachers to understand the prevalence and impact of NCII in K-12 schools.
The findings indicate that NCII is a significant issue in these educational settings. Both students and teachers reported substantial amounts of authentic and deepfake NCII being shared, with students often being both perpetrators and victims. Female and LGBTQ+ students are particularly affected by this issue. They expressed lower confidence in their schools' ability to prevent and respond to deepfake NCII.
"Students and teachers report that female students are more often depicted in deepfake NCII that is shared by their classmates," the report states. Additionally, both female and LGBTQ+ students indicated they have less faith in their school's preventive measures against such threats.
The report also points out that many schools lack proactive policies to address the spread of NCII. Instead, they tend to react only after an incident occurs. "Very few teachers report that their schools have policies and procedures that proactively address the spread of authentic and deepfake NCII," according to the CDT's findings.
When incidents do occur, schools focus heavily on penalizing perpetrators through measures like expulsion or long-term suspension but provide limited support for victims. "Both students and teachers say that schools provide few resources for victims of NCII," such as counseling or assistance with removing harmful content from social media platforms.
Parents appear less aware of these issues compared to students and teachers but agree on the need for better education on this topic. They believe they should play a primary role in educating children about these risks.
The CDT concludes that addressing both authentic and deepfake NCII will require a long-term approach involving multiple stakeholders. However, immediate steps can be taken by schools to bolster prevention measures, improve victim support, and engage parents more effectively.
For further details, readers are encouraged to read the full report or explore the accompanying slide deck.
