Skip to content
Submit incident
AII-1002-02723
ReportedReported by fewer than three independent publishers, with no official finding.

AI-Generated Deepfake Images Used to Sexually Exploit Spanish Minors

A Save the Children report found that 20% of Spanish young adults surveyed reported being victims of AI-generated deepfake sexual images during their childhood, with 98% having experienced some form of digital sexual violence as minors. The report warns that AI-generated deepfakes of minors pose a "new challenge of enormous gravity" for child protection, as images from social media can be easily converted into material for exploitation.

7 July 2025 · Harmful and abusive content
7 July 2025
AII-1002-027232 sources ↓
Quick facts
Category
Harmful and abusive content
Country
Spain
Source
LaVanguardia
Date
7 July 2025
Drafted by
Raidu Incident Detection

What happened

A report by Save the Children, published in collaboration with the European Association for Digital Transition, surveyed over 1,000 young people aged 18-21 in Spain about digital sexual violence during their youth. The study found that 97-98% of respondents reported experiencing some form of digital sexual violence when they were minors, including contact by adults for sexual purposes, non-consensual sharing of intimate content, and creation of deepfake images.

According to the report, 20% of the surveyed young adults reported being victims of AI-generated deepfake sexual images that showed their bodies as naked, created without consent and distributed. The report notes that generating this type of content can be accomplished using images readily available from social media, as creators need multiple angles of the victim that are often publicly posted.

Reported impact

The report characterizes digital sexual violence as 'massive, diverse and expanding' and identifies AI-generated deepfakes as 'a new challenge of enormous gravity' that can convert any image into 'raw material for exploitation.' The combination of real abuse images with digitally generated content complicates the work of law enforcement in distinguishing between actual victims and AI-generated content.

The report notes that 70% of surveyed young people do not perceive manipulation of photos or videos using AI as a risk, indicating a significant gap in awareness about this threat among potential victims.

Index analysis

The exploitation takes advantage of the widespread practice of parents and guardians sharing photos of children on social media. These publicly available images can be harvested and used without consent to create deepfake sexual material, making any child with an online presence a potential target.

Governance controls

The Spanish Government has promoted legislation establishing penalties of up to two years imprisonment for falsifying sexual images using AI.

Classification

Category
Harmful and abusive content
Industry
Media, social platforms and marketing
Country
Spain
Harm status
Actual harm
Intent
Intentional
Timing
Unknown
Harm types
Psychological, Human or fundamental rights, Public interest
Harmed
Children
Failure modes
Deepfake image
Consequences
Legislative action

Sources and evidence

This record was researched and written by the Index, based on the following original reporting.

LaVanguardia
Alerta ante el uso de la inteligencia artificial por los depredadores sexuales de niños
7 July 2025
20 minutos
El 98% de los jóvenes dice haber sufrido violencia sexual digital y uno de cada cinco ha sido víctima de 'deepfakes' con IA
8 July 2025